The qualitative characterization of Sangiovese grapevine ... · Ph.D thesis IN SCIENCE OF CROP...

Ph.D thesis

IN SCIENCE OF CROP PRODUCTION

XXIV Cycle (2009-2011)

AGR/03

The qualitative characterization of

‘Sangiovese’ grapevine according to

the area and cultivation conditions

CANDIDATE Eleonora Ducci

TUTORS COORDINATOR Prof. Giancarlo Scalabrelli Prof. Alberto Pardossi

Dr. Claudio D‟Onofrio

November 2013

Department of Agriculture,

Food and Environment

(DAFE)

University of Pisa, Italy

Life is wonderful all you need is to look at it with the

‘right eyes’

TABLE OF CONTENTS

1. INTRODUCTION .................................................................................................... 1

1.1 Sangiovese cultivar ................................................................................................. 1

1.1.1 Adaptability of the soil ................................................................................. 5

1.1.2 Adaptability to temperatures ......................................................................... 6

1.1.3 Adaptability to rootstock .............................................................................. 7

1.1.4 Adaptation to the training system and density of plantation ........................ 8

1.1.5 Adaptability to summer pruning ................................................................... 9

1.2 The terroir ............................................................................................................. 12

1.2.1 The terroir and its evolution ....................................................................... 12

1.2.2 The vineyard factors ................................................................................... 16

1.2.3 Terroir of Tuscany ...................................................................................... 19

1.3 Aim of the thesis ................................................................................................... 22

2. MATERIALS AND METHODS ........................................................................... 23

2.1 Plant material ........................................................................................................ 23

2.2 Technological maturity ......................................................................................... 25

2.3 Sensory analysis on grapes ................................................................................... 25

2.4 Phenolic compounds analysis ............................................................................... 27

2.5 Aroma compounds analysis .................................................................................. 27

2.5.1 Preparation of grape sample ....................................................................... 28

2.5.2 Enzymatic hydrolysis of glycosides ........................................................... 28

2.5.3 Acid hydrolysis of glycosides ..................................................................... 29

2.5.4 Gas chromatography – mass spectrometry ................................................. 29

2.6 Statistical analysis ................................................................................................. 30

3. RESULTS ................................................................................................................ 31

3.1 Weather station location and study of the historical sequences ............................ 31

3.2 Climatic characteristics ......................................................................................... 35

3.3 Vineyards‟s characteristics ................................................................................... 43

3.3.1 Chianti Colline Pisane ................................................................................ 43

3.3.1.1 Beconcini estate ................................................................................ 43

3.3.2 Montecucco ................................................................................................. 44

3.3.2.1 ColleMassari estate ........................................................................... 44

3.3.2.2 Salustri estate ................................................................................... 47

3.3.3 Morellino di Scansano ................................................................................ 48

3.3.3.1 Fattoria di Magliano estate ............................................................... 48

3.3.4 Brunello di Montalcino ............................................................................... 49

3.3.4.1 Col d‟Orcia estate ............................................................................. 49

3.3.4.2 La Mannella estate ............................................................................ 49

3.3.4.3 Casanova di Neri estate .................................................................... 50

3.3.5 Chianti Classico .......................................................................................... 51

3.3.5.1 Castello di Albola estate ................................................................... 51

3.3.5.2 Capannelle estate .............................................................................. 51

3.4. Soil‟s characteristics ............................................................................................ 52

3.5 The study of geopedologic, orographic and viticultural aspects: the ColleMassari

estate .................................................................................................................... 56

3.5.1 Campo La Mora ......................................................................................... 58

3.5.2 Cerrete ........................................................................................................ 60

3.5.3 Orto del Prete ............................................................................................. 63

3.5.4 Vigna Vecchia ............................................................................................ 65

3.6 Harvest date .......................................................................................................... 69

3.7 Sensorial characteristics of the grapes at harvest‟s time ...................................... 70

3.7.1 Montecucco area ........................................................................................ 86

3.7.2 Col d‟Orcia estate ..................................................................................... 103

3.8. Technological and phenolic characteristics of the grapes at harvest‟s time ..... 119

3.8.1 Montecucco area ...................................................................................... 131 3.8.2 Col d‟Orcia estate ..................................................................................... 143

3.9 Aroma characteristics of the grapes at harvest‟s time ....................................... 156

3.9.1 Montecucco area ...................................................................................... 173

3.9.2 Col d‟Orcia estate ..................................................................................... 184

4. DISCUSSION AND CONCLUSIONS ............................................................... 195

5. REFERENCES ..................................................................................................... 202

1

1. INTRODUCTION

1.1 Sangiovese cultivar

Sangiovese is the base vine variety of Tuscan enology as it is the main component of the 7

Tuscan DOCG, since its presence varies from a minimum of 50% to a maximum of 100%:

„Brunello di Montalcino‟ (100%), „Carmignano‟ (50%), „Chianti‟ (70-100%), „Chianti

Classico‟ (80-100%), „Morellino di Scansano‟ (85%), „Montecucco‟ (90%) and „Nobile di

Montepulciano‟ (70%). It has many synonyms, the official one reported in National Register

of the Varieties of Vines, „Sangioveto‟, and others certified as „Brunello‟ (Tuscany),

„Morellino‟ (Scansano-Grosseto), „Nielluccio‟ (Corse, FR), „Prugnolo‟ (Tuscany), „Prugnole

gentile‟, „Sangiogheto‟, „Sangiovese grosso‟, „Sangiovese piccolo‟, „Sangioveto montanino‟,

„San Zoveto‟, „Uvetta‟ (www.vitisdb.it). Reconstructing the origin of this vine is not easy

since there is a lack of historical reference antecedent the XVI century. The importance of this

variety in wine production in central Italy and its leading role today in Italian enology justifies

the interest in searching the origin of its name. Due to lack of accurate references it was first

thought that it recalled the idea of blood, one of the symbols linked to wine and to offering

sacrifices to the gods, the blood of Jove (sanguis Jovis). The semantics of the word recalls a

game (jugum), and sustains the hypothesis of sangue-gio-vese, hill games; another hypothesis

is that of wine „giovevole al sangue‟ (Mainardi 2001). Other connections have been

hypothesized with language and popular customs, between the Etruscan language, religious

aspects and the meaning of the term „Sangiovese‟. An Etruscan phrase was found on a

bandage used to wrap an Egyptian mummy of the first century AD, it read „s‟antist‟celi‟ with

the word „vinum‟ and it is thought that it referred to a type of wine as it is very close in

assonance to terms that describe „Sangiovese‟. Other assonances linked to rituals and

„Sangiovese‟, as in thana-chvil (votive offering), tbcms.zusleva (ritual offering), thezin-eis

(offering to the god), which is very close to the Romagnolo term sanzve used for

„Sangiovese‟, this term means father or ancestor to mean the wine of my father‟s or an

offering to the fathers (Mainardi, 2001).

Linking the origin of the Sangiovese vine variety to the Etruscan culture is fascinating, recent

discoveries, the close relationship between „Ciliegiolo‟ e „Calabrese di

Montenuovo‟(Vouillamoz et al., 2007) and (Bergamini et al., 2012), do not concord with

these hypothesis even if they do not totally negate them as shown in other research (Di Vecchi

et. al, 2007). A current hypothesis associates the name of the vine to „sangiovannese‟ as in

2

originating in „San Giovanni Valdarno‟, others believe the origin of the name to come from

dialect „sangiovannina‟ early grapes. The first reference to the existence of this vine variety in

Tuscany dates back to Soderini (1590) who calls it „Sangiogheto‟. At the end of the sixteenth

century it appears in a painting by Bartolomeo del Bimbo known as „il Bimbi‟ by the name

„Sangioeto‟ (Basso, 1982), while Trinci (1738) describes the „SanZoveto‟ as „a fine quality

grape and bountiful in production‟. Moreover the reliable productivity characteristics of the

„San Gioveto‟ are praised by the Villifranchi in his „Oenologia Toscana‟ (1773) defining it

„the protagonist of Tuscan wine superb in taste and generous‟. Villifranchi (1773) also refers

to „San Gioveto‟ strong (synonymous of dog deceiver) and „San Gioveto romano‟ that is

cultivated in Marca and in particular in the Faentino area where the wine produced is

generous „call it San Gioveto‟. The existence of the „Sangovese‟ wine and the description of

its qualities are found in convivial texts and in the dithyramb of 1818 „Il Bacco in Romagna‟

by the abbot Piolanti (Mainardi, l.c.).

From Tuscany and „Romagna‟ the elected areas the cultivation on „Sangiovese‟ spread

progressively to other Italian regions, „Marche‟, „Umbria‟, „Abruzzo‟, „Lazio‟ and „Puglia‟

(Mainardi l.c.) and „Corsica‟. Most of this growth occurred at the end of the nineteenth

century and the beginning of the twentieth century with the reconstruction post phylloxeras.

A large scale renovation of the plants took place in the 60s and 70s thanks to the „Piano

Verde‟, that gave incentives to expand vineyards. The setting for quantity and choice of plant

sites which were not always the best did not help the development of this vine variety, but

have indeed limited it. The obsolescence of these vineyards has requested renovation, paying

particular attention to choice of soil, cloning material and plant design. The latter has been

orientated towards an increase in plant density and the accomplishment of management

techniques to obtain high quality grapes, able to produce important red wines. (Loreti and

Scalabrelli, 2007). At present the „Sangiovese‟ vine is the most diffused in Italy, and

according to the ISTAT (General Census Survey), in the year 2000 about 70.000 hectares

were cultivated covering over 10% of the total surface in vineyards; this data is also

confirmed in the 2010 statistics. In Tuscany it is the most diffused vine variety, covering

37.170 ha 67,4% of the regional viticulture surface. As for the agronomic properties, the

„Sangiovese‟ variety is characterized by a rather early bud burst, found along the Tuscan

coastal areas in the last 10 days of March and about a week later in the inner areas. This vine

needs high temperatures for ripening (Turri and Intrieri, 1988), reaching its peak around the

last 10 days of September in the coastal areas while inland and hilly areas early to mid

October. Adaptation to colder climes is linked to rainfall in the month preceding harvest. The

3

high fertility of the base gems accounts for the spur pruning which can be very short in the

hottest areas („Montalcino‟, „Maremma‟). Many types of training system can be practiced,

short pruning (tree-like, spur pruned cordon, GDC), mixed (Guyot, „capovolto‟), long

(„tendone‟): these are chosen on the basis of climatic conditions and soil fertility. Rootstocks

are now more used than in the past in areas where there are no risks of prolonged drought,

high density plantations are employed choosing less vigorous subjects (161/49,101,14), to

110R where there is a need for more drought tolerance, while in the most difficult conditions

1103P is used.

Figure 1. Sangiovese‟s bunch.

It is greatly adaptable to diverse environments, even if in coastal areas it can suffer from late

frosts. High quality grapes are produced in low fertility soils, well drained and dry climes,

with moderate lack of water from veraison to ripening. For a better aroma complexity it is

also important to have good temperature range. The terroir effect is well shown by the

particular characteristics of wines from different areas. As already stated the „Sangiovese‟ is

the base vine variety in Tuscan oenology besides being the king of wines like „Chianti‟,

„Brunello di Montalcino‟, „Nobile di Montepulciano‟ and „Morellino di Scansano‟ (fig. 2), it

is the main vine variety in the production of almost all red DOC and IGT wines in Tuscany.

The bunch (fig. 1) is average size, conic in

shape and average compactness.

The sugar levels reached in the right

conditions is high, while the anthocyanins

content of the skins is greatly influenced

by the site, the cultivation technique and

in particular by the vigour.

The different clones offer a variety of

choice according to the morphology and

the qualitative characteristics of the bunch

(Moretti et al., 2007; Tamai, 2009), so as

to allow the realization of polyclonal

vineyards.

4

Figure 2. Consortium Logos of some DOCG produced with Sangiovese.

Among the DOC there are „Barco Reale di Carmignano‟, „Bolgheri rosso‟, „Candia dei Colli

Apuani‟, „Capalbio‟, „Colli dell'Etruria Centrale‟, „Colli di Luni‟, „Colline Lucchesi‟,

„Cortona‟, „Elba‟, „Montecarlo‟, „Montecucco‟, „Monteregio di Massa Marittima‟,

„Montescudaio‟, „Orcia‟, „Parrina‟, „Pietraviva‟, „Pomino‟, „Rosso di Montalcino‟, „Rosso di

Montepulciano‟, „San Gimignano rosso‟, „Sant'Antimo‟, „Sovana‟, „Terratico di Bibbona‟,

„Val di Cornia‟, „Valdichiana‟, „Vin Santo Occhio di Pernice‟. Among the IGT „Sangiovese‟

is among the components of: „Alta Valle del Greve‟, „Colli della Toscana centrale‟,

„Maremma Toscana‟, „Montecastelli‟, „Toscana‟, „Val di Magra‟.

„Sangiovese‟ is also used in the production of DOP and IGP wines in other regions too;

„Bardolino‟, „Garda Est,‟ „Valdadige,‟ „Valpolicella‟, „Sangiovese di Romagna‟,

„Montefalco‟, „Rosso piceno‟, „Rosso Conero‟, „Velletri‟ and „Gioia del Colle‟.

Depending on the area, grape characteristics and level of phenol ripening, it is possible to

obtain rosé wines, young red wines ready to be drunk and wines suited to short, mid or long

maturation. One of the problems with „Sangiovese‟ is linked to the quality of the grapes

which are heavily dependent on the climatic course of the year. The grapes can be turned into

wine in blend with other vine varieties according to the objectives desired. Grapes that are in a

5

good healthy state give a tannic product needing refinement before it can be consumed. The

colour stability greatly depends on the anthocyanic composition, that in the „Sangiovese‟ is

not optimal for lack of malvidin; however this problem has been mitigated improving the

production techniques (minor yield per plant) and by using qualitative clones. „Sangiovese‟ is

also a blend vine, as shown by the formula in „Chianti‟ del „Barone Bettino Ricasoli‟ (7/10 of

„Sangiovese‟, 2/10 od „Canaiolo nero‟ and 1/10 of „Malvasia bianca lunga‟), and it has

evolved from being a year wine to refined wine with the progressive reduction of white berry

vines. The red berry vines used in the blend are there to integrate the characteristics of the

„Sangiovese‟ wines in particular years or in less favourable conditions to give greater colour

stability, greater sense of smell and mellowness. The young wine is an intense rich red colour,

red fruit scented and at times floral and or vegetable, dry tasting, correctly tannic. Wines that

are to be refined are more structured and have a higher acidity level. With ageing the colour

tends to garnet and with the fruity note there are the evolved scents of tobacco, balsamic and

liquorices.

1.1.1 Adaptability of the soil

Tests conducted on the influence of the soil on the quality of „Sangiovese‟ grapes in the

„Chianti Classico‟ area, have shown a link between the sugar content of the grapes and nature

and soil composition and in particular the organic and clay content. The best soils are those

with average fertility, clayey-chalky and well framed, that dry quickly during ripening and as

such in these soils the vegetative development of the plants is more balanced. According to

Bertuccioli (2000) the most interesting values of the chemical parameters linked to quality,

were found in areas with a higher percentage of sand and with a lower rate of phosphorous

and potassium that can be assimilated. Tomasi and others (2006), have pointed out that in

cases of grapes from non chalky soils, in the corresponding wines there was a strong spicy

scent, cinnamon and cherry, but above all they presented a fullness of taste that was not

present in chalky soils. In these soils however, the aromatic fineness and persistence

triumphed, and the scent of violets and white flowers. At high temperatures the monoterpenic

substances are lower. The reduction of aromatic content of the grapes, due to high

temperatures, is so strong so as to also cancel the positive action linked to water content of the

soil. It was also possible to note the norisoprenoids components present in the grapes grown

in quite damp soils. The high temperatures therefore compromised the aromatic quality,

independently from the water content in the soils.

6

From a survey carried out by the CRA-VIT (Sebastiani and Storchi, 2004) on a number of

vineyards in the „Arezzo‟ province, the influence of soil management relative to a number of

vegetal-productive plant response emerged. Therefore, the positive response of „Sangiovese‟

to grass cover became obvious, even if limited to alternate rows. In particular, with this

technique the result was a lower average weight of the bunch and of the berries, but with

positive effects on sugar content and colouring substances and on the state of health of the

grapes at harvest. All this had already been noted by Egger et al., (1996). These results have

also been confirmed in other research by Bertuccioli et al., (2000) carried out on the grass

cover in two areas of the „Chianti Classico‟. The wines from grass covered vineyards have

clearly shown the positive influence of this technique on the quality of the product having

higher alcohol content and greater net extract, total polyphenols and anthocyanins compared

to vineyards situated on land worked so as to support the vigour of the vine (Pisani et al.,

2000). Triolo and Materazzi (Triolo et al., 2000) too have noted how grass cover favours

significant reduction in Botrytis, particularly in years of low rainfall.

1.1.2 Adaptability to temperatures

Intrieri, already in the 1980s, underlined the resistance to the cold of the main buds of the

„Sangiovese‟. After the bad frost in January 1985 at -18°C, the bud mortality was just over

20%,but with the further fall of 1°C the mortality rose to 90% thus setting the critical

threshold to the winter cold at an interesting -18°, -19°C. In tests in the 1970s the

environmental stability in the phonological phase was evaluated and variable behaviour per

bud, flower and veraison was observed in that the vine often suffers environment conditions

but not always in an univocal way (Calò et al., 1977). Test Results carried out in 2004 have

underlined the importance of high thermal summing for the perfect completion of the

vegetative cycle and that the best quality is determined by temperature along with water

availability in the fruit set- veraison period, so much so that the best years are correlated to

higher average temperatures summing and lower rainfall values in the vegetative period. In

the „Montepulciano‟ area there was a positive correlation between the altitude of the

vineyards and malic acid content in the must, as late harvesting occurs in altitude and this

confirms the temperature needs of this vine variety during ripening (Egger et al., 1986).

Systematic research conducted since 1987 on the relationship between variety and

environment, has shown that the „Sangiovese‟ is more reactive to pedoclimatic and cultivation

solicitations. Results from four different years (1987-1990) from the main vine growing areas

7

of Tuscany („Chianti Classico‟, „Montalcino‟ and „Montepulciano‟), show higher sugar levels

in areas where the mean temperatures are higher and rainfall levels are lower during the

vegetative period. The values obtained from the Huglin index, on the other hand, appeared

less correlated to the sugar content at harvest. In the „Montepulciano‟ area there was a

positive close correlation between the altitude of the vineyards and the malic acid content and

the titratable acid of the must. In relation to altitude the ripening period too is later due to the

lower temperature levels in the period preceding harvest. Research in the „Chianti Classico‟

area clearly show the vine sensitivity to rising temperatures. In particular, comparing the two

maturation curves obtained from vines with the same productive weight but in different

altitude environments (600 and 350 m. above sea level) showed for the latter a constant higher

sugar level (Scalabrelli et al.,1996). This difference is to be attributed presumably to the

beneficial effects that rising temperatures together with periods of sun exposition, can have on

the total photosynthetic yield of the foliage as a consequence and the ability to accumulate dry

substance in the bunches.

Bunch size is determined by climate and temperature and is one of the determining factors in

the synthesis processes, accumulation and conservation inside the berry of the aromatic

mixtures. Another important positive link is between climatic parameters and aromatic

substances between the norisoprenoids content and potential value of photo- synthetically

active radiation. It is necessary to add the effect of the active limestone content in the soils, in

fact in equal active photo synthetic radiation the norisoprenoids content is distinctly higher in

vineyards with greater limestone content (Failla, 2006).

1.1.3 Adaptability to rootstock

The adoption of rootstock suitable in different ecopedologic and cultural conditions and

productive typologies is of crucial importance in order to obtain the qual-quantitative results

desired. Even in this sector there is much reliable data from research carried out on the

„Sangiovese‟. In the area of the „Morellino di Scansano‟, Di Collato et al., (2000) noted the

tendency of higher sugar concentration levels in grapes in vines grafted on 110R followed by

SO4 and 41B. A good hold on acidity levels was observed in combinations with rootstock 140

Ru, 3309C and 41B, while in most of the other thesis the values registered were much lower.

Taking into account the generalized tendency to the decreasing productive yields in the areas

of Denomination of Origin, where red wines are produced, it is worthwhile noting that lower

productivity induced by 41B, differently from other rootstocks, determined some of the

8

highest sugar gradation, while keeping a good hold on acidity levels. In the „Chianti Classico‟

the experimental trials carried out on rootstocks by Scalabrelli and Loreti (Scalabrelli et al.,

2000) evidenced the different influence on „Sangiovese‟ both in the vegetative-productive

performances and in qualitative aspect. On the basis of the results the SO4 is not the ideal

rootstock for the pedoclimatic characteristics of this area. With this rootstock, the vines often

produce an excessive quantity of grapes with a mediocre sugar level.

The rootstocks by V. Berlandieri x V. Rupestris (775P, 779P, 1103P), even if producing a

certain vigour and productivity, have also determined a reasonable sugar gradation of the

grapes. However, some years it has been necessary to reduce the quantities produced. The

140Ru, 110R, 225Ru, 420A, 34EM, rootstocks have induced average to low vigour and quite

good sugar gradation and production. Rootstocks 101.14 and 3309C have induced lower

vigour, higher sugar gradation and relatively low productivity even lower than the regulatory

foreseen for the „Chianti Classico‟ DOCG. From this we can ascertain that the 110R is the

best rootstock for the global performance of the „Sangiovese‟ in the „Chianti Classico‟ soil

conditions Scalabrelli et al., l.c.).

1.1.4 Adaptation to the training system and density of plantation

Intrieri points out „the high fertility of the basal buds of the „Sangiovese‟ vine shoot offers a

wide range of choice in terms of pruning lengths and consequently training systems. On this

premise the author carried out many tests to observe the behavior of different systems with

long pruning with annual renewing of shoots (Guyot unilateral or bilateral) and with

permanent cordon with long pruning („Casarsa‟ type) or short (spur pruned cordon type,

single and double T, short and long GDC, single „Cortina‟) (Filippetti et al., 2000. Intrieri et

al,. 1985, 1992, 1993, 2000).

In the results the „Sangiovese‟ has always shown a tendency to stay on high unitary yield

levels and above all the remarkable capacity for productive compensation, if blocked on

rather modest bud weight. From the systems observed, only the long GDC and vertical

cordon have resulted in lower than the average production remembering that this system (in

the case of GDC) lowers the vigour. In the vertical cordon, acrotonic gradient has brought

physiological imbalance, shading from the excessive bud growth and resulting in product

penalization. The result highlighted even by the author of the test, is that the „Sangiovese‟

vine is adaptable to training systems that can be diversified per vine structure, bud weight and

crown architecture.

9

Varying these conditions, the vine maintains a strong productive capacity with appreciable

qualitative levels. However there is also the need to find systems that can contain the natural

tendency of the vine to let the vegetative phase prevail; this phenomenon has also been found

in other tests by Intrieri et al., 2000.

Bertuccioli et al., (2000), in a study, noted that average-low density plantation determined a

likely physiological imbalance of the vines in favour of vegetative growth, penalizing the

quality of the production.

Moreover, according to Bertuccioli (2000), plantation density, showed significant effects on

the quality of the „Sangiovese‟ grapes. In particular, the most interesting values of the

chemical parameters linked to quality have been observed, in the area characterized by a

higher percentage of sand and lower values of phosphorous and potassium that can be

assimilated and by the catatonic exchange in the higher densities, while on the richer soil they

coincide with the average-low densities, which as a result of the excessive soil exploitation

determines a physiological imbalance of the vines in favour of vegetative growth.

Scalabrelli, et al., (2000) have noted that even in diverse environments and wine typology,

density of around 5.000 stumps per hectare in general determines a better balanced plant

growth. Higher density do not offer advantages, that can be generalized, on the vegetal-

productive behavior and on the quality of the grapes. Indeed, depending on the ecopedologic

conditions problems can arise due to the alteration of the vegetative balance if a low number

of buds per plant is chosen, or production increase if buds with a greater weight are chosen.

1.1.5 Adaptability to summer pruning

Shoot thinning the vine during the vegetative cycle, is quite diffused and this has posed and

poses many physiology problems, particularly concerning the flow of the elaborates at the

time of intervention.

Several works conducted by Calò (1975-1976) clarified the balance in the vegetative and

accumulation phase, and how it can be compromised in relation to trimming, which cannot be

considered as the operation that reduces the surface foliage of the plant.

In some studies by Intrieri and collaborators (Intrieri et al., 1983, 1985) confirmed that good

results were obtained by trimming 12 days after flowering. By stimulating the development of

the laterals shoots able to reach physiologic ripening at veraison and thus in time to contribute

to nourishing the bunch, has given rise to a good level of ripeness. On the other hand, a late

10

trimming, that is at the time when the push in the development of the shoots is inferior, has

brought about a slowing down in the growth of the berries and their ripening.

Experiments by Palliotti (2000) the trimming of the shoots carried out prematurely has

resulted in the early development of the shoots, in the building sufficient surface foliage

sufficiently to guarantee in an optimal way the ripening processes of the grapes, without

modifying vine productivity and improving grape quality. Various factors contribute in

reaching such results: foliage reduction of the crown resulting in better light penetration; to

the source of the shoots thereby improving light intensity and the radiation red ratio far away

in the grape area, rejuvenating the crown with early shoots whose leaves have shown since

veraison up to their abscission, higher photosynthetic activity than the main leaves. This has

drastically reduced foliage surface necessary to bring into full maturation the weight unit of

the grapes. After defoliation, Ferrante et al., (1999) did not observe production compromise

neither in qualitative nor quantitative levels, not even with intensity over 60 %. Late trimming

caused the slowing down of the sugar accumulation and the acidity degradation, thereby

delaying the best time to harvest and induced product quality deterioration. These results

backed by other authors, are probably due to the late trimming of the vines, laterals shoots

development continues even after veraison, thereby reducing sugar in the bunches. This new

vegetative development delays the accumulation phase and consequently the best harvest

time. Another in summer pruning studied was defoliation. In 2012 studies were made by

D‟Onofrio et al. on the effects of defoliation carried out in different periods, on the

characteristics of the berries and on the aromatic quality of the „Sangiovese‟. These

experiments reported how the berries from the non defoliated sample had a higher greater

berry weight compared to the two thesis defoliated at pre-flowering and ay veraison. With

reference to the aromatic component of the grapes a progressive accumulation of diverse class

aromas reaching peak point, from which it starts decreasing up to harvest time. Such a

decrease is probably due to the degradation of the aromatic composites induced by the high

temperatures recorded in the last phases of ripening.

The fruit set defoliation determines an increase in content of aromatic composites, compared

to the non defoliated sample. A significant increase was noted in the monoterpenols and C13

norisoprenoids, a slight increase in benzene derivatives and a decrease in the aliphatic

alcohols. The veraison defoliation thesis, presented on the other hand, aromatic composites

content inferior to the non defoliated control thesis. Early defoliation is therefore an important

and effective in increasing aroma concentrations and, as a result for the aromatic quality of

the grapes. The same vineyard was monitored in 2009. This time the thesis analyzed were

11

two: non defoliated control and fruit set defoliation. In the defoliated sample an increase was

noted in all the composite classes, above all at the expense of the C13-norisoprenoidi,

exception made for aliphatic alcohol class. The experiment was also carried out the following

year with the addition of a thesis created by covering some bunches. It was highlighted how

the defoliated thesis at flowering and the fruit set defoliated thesis, had at harvest a higher

concentration of all the aromatic composite classes studied. Among these the greatest increase

was reached in the fruit set thesis where there was a significant increase in the concentration

of all the composite classes. In the defoliated at flowering thesis and later covered it was

noted that the bunch temperature was about the same as the temperature of the defoliated at

flowering thesis and therefore the only variable, at the base of the significant reduction of

monoterpenols and C13-norisoprenoids concentrations, can be attributed to the absence of

light.

Reduced leaf layer numbers in a vine may have many beneficial impacts greater anthocyanins

and phenolics in Sangiovese harvested after leaf removal (Poni et al., 2006).

In 2008 Intrieri and Filippetti, compared manual and mechanical defoliation on a

„Sangiovese‟ cultivar: the first six basal leaves and any laterals were removed by hand, and

the same area was subjected to mechanical defoliation, the latter removing 48.3% of the leaf

area removed manually. Both treatments significantly reduced fruit-set, yield per shoot, bunch

weight, berries per bunch and bunch compactness. Yield/ha declined from 32.8 tons in control

vines to 24.4 and 19.0 tons for mechanical defoliation and hand defoliation (pre and post

bloom treatment means), respectively. Leaf to fruit ratios were unaffected by defoliation as

source loss was fully offset by yield decline. Soluble solid concentration and total

anthocyanins on a fresh-weight basis increased by 2.4°Brix and 0.2 mg/g in hand defoliation

and by 2.2°Brix and 0.08 mg/g in mechanical defoliation as compared with that in non-

defoliated control. Although results from hand defoliation reinforce the physiological basis of

the technique‟s effectiveness, mechanical defoliation proved likewise effective in reducing

yield and improving grape quality.

12

1.2 The terroir

Even in well known and prestigious „denomination of origin‟ vine growing areas there are

different vine behavior patterns and therefore the grape and wine characteristics in time and

area. It is a well known fact that soil, climate, vine variety and cultivation technique are the

main factors in influencing the productive and qualitative result of the vines.

Appraising the cultivation vocation of the territory is one of the best instruments in

safeguarding the typicality of the products and the risks involved in soil degradation. In

particular, studies on the correlation between the quality of the environment and the quality of

the product show the good use of local resources which is strictly connected the specificity of

the environment of origin and that must be safeguarded. Such specificity is usually known by

the term „cultivation vocation‟ and this is regarded as one of the most important success

factors in national agriculture in the global market because very often quality cultivation

becomes a reference point and a leading image of the territory. Indeed, the term „total quality

of the territory‟ refers to territory managed in function of product quality, soil and ecosystem

conservation, healthy environment and landscape beauty. The acknowledgement of the

„vocation‟ of a territory is needy of research into its peculiarities which exalt its

„exclusiveness‟. In other words, it is the peculiarities of a territory and its functionality, the

influence that determines the variability of response, the quality of a wine or of an olive oil

for example, that determines the uniqueness of that particular production area. The uniqueness

of a production area is thus and added value to the quality that can be crucial for the success

of a cultivation. The distinguishing characteristics that determine a production area suited to

quality food producing are better explained and detailed in the single functional components

of the territory, the „terroir‟.

1.2.1 The terroir and its evolution

The French term terroir which is difficult be to translated in other languages it nowadays

utilized in the wine world communication, its meaning can be described as a complex

combination of factors that determine a specific wine characteristics not repeatable elsewhere.

Regarding the roots of this term it is useful to remember that over a century ago its meaning

was very different, as revealed by Lawely (1870): „it is to be remembered that our hills

created by terrestrial clay-limestone white grapevine varieties perform very well because

13

they can acquire a perfect maturity, that elsewhere it is not possible: cultivated black grapes

in that soil could equally produce good wines , if these didn't receive from the soil a

particular taste of terrain, that French call terroir‟.

The term terroir has since been modified and enriched with other meanings, especially in

consequence of the intensification of viticulture and wine research and for the its wide use

by producers, researchers, journalists, wine critics and wine consumers. Today this word

contains at least the following four specifications Origin, Specificity, Perennial and Typical

(fig. 3).

Terroir:Origin,

Specificity, Perennial and

Typical

AGRICULTURAL: Terroir-subject

TERRITORIAL:Terroir-space

IDENTITY:

Terroir-conscience

ADVERTISING: Terroir-slogan

Figure 3. The multiple meaning of Terroir (Scalabrelli, 2013).

The AGRICULTURAL or Terroir-subject is identified by the Taste (sensory characteristic)

and from the consequent Instrumental Quality, the complex features due to relationships

between plant and environment. The TERRITORIAL concept or Terroir-space, concerns the

delimitation of a territory, with the relative Denominations and their specificities recognised

by the Unity of landscape that are also referred to as historical Geography of the territory.

The terroir IDENTITY or Terroir-conscience, constitute the immaterial part that is present in

a determined country, represented by identity and by the sense of affiliation of the inhabitants,

in relationship to their genealogical origin and their traditions.

The ADVERTISING terroir or Terroir-slogan constitutes the communicative part of the rural

world which needs to express certain values and to transmit one specific image. This means

searching meanings that are patrimony of the producers and that must be rendered explicitly

well through communication in order to offer a better understanding of the essence and the

specificity of the viticulturists job.

14

The terroir can be considered a complex system where the genotypes (rootstock, variety and

clone) interact with the soil, the macro-meso-climate (Carbonneau, 2000) and the viticulture

model (planting design and density, training system, geometry and canopy extension)

determining a functioning ecosystem, modulated by techniques of management (fig. 4).

Terroir:Wine composition

and qualityattributes obtained

by FactorsInteraction

driven by the man

Genotype: Rootstock,

Variety and Clone

Macro-meso-climate

and TerritoryEcosystem

Wine Technology:

Valorization ofgrapes features,

and wine process

Soil: Origin, Structure,

Fertility, Depth, Physical,

Chemical and Microbiological Characteristics

Viticulture model: Planting Design, Training System, Canopy

Geometry

Figure 4. Factors involved into the Terroir effect (Scalabrelli, 2013).

The interaction genotype environment, influences the vine sink/source relationships and, in

general, the production efficiency of the canopy. Therefore the quality and the composition of

grapes depend on vine equilibrium, or rather from the correct relationship between the leaf

functioning area and the amount of yield (Casternan, 1971; Scalabrelli et al., 2001; 2003;

Fregoni, 2005).

The knowledge of the vocation of the territory is acquired through interdisciplinary studies of

zoning, that through an integrated approach (Morlat et al., 1989) they aim to understand the

mechanisms of interaction environment x macro-meso-climate that affect the grapevine

physiology (Asselin 2001, Asselin et al., 2003). The system terroir/vine/wine can be

represented by a pyramid constituted by variables of simple and composite state, parameters

of functioning, and variables of operation, vintage and wine. All these aspects, in relationship,

define the system terroir/vine/wine (fig. 5).

15

Wine

Harvest

Variabiles of Functioning

(earliness, water nutrition..)

State Composed Variables

( permeability, porosity)

State Simple variables

(altitude, mineralology, granulometry,

chemical composition..)

Characteristics

of

the

production

Plants Characteristics

Environmental

features

Parameter of functiong

( pedoclimate, mesoclimate)

Figure 5. Chain of influences involved in the Terroir effect. Redrawn from Asselin (2001)

and Scalabrelli, (2013).

Hence this system implicates the impossibility to expound the elaboration of the wine quality

and the operation of the terroir through the simple measurement of the influence of a single

factor (Morlat, 2001).

Through an integrated approach it was initially defined the scientific concept of terroir

(Asselin, l.c.) as the basic unit terroir (UTB), which can be defined as „the smallest vineyard

area used by the grower in which the response of the vine is reproducible through the wine‟,

or „the smallest physical unit homogeneous that we can usefully differentiate, for practical or

for scientific purposes’, which is a separate operating unit agro-ecosystem „physical site x

vine’ (Riou et al., 1995). The unit of viticulture terroir (UTV) is the smallest unit that can be

formed by UTB, locally defined (Carbonneau, 1993) with the variety, cultivation techniques

and wine (Deloire et al., 2002). Moreover when the base of the UTV group are identical or

similar and associated with a strong personality of the wines (sometimes without complexity),

they give rise to a homogeneous viticulture terroir. If several UTV are different, they form a

„composite terroir’. In this case the personality of the wines is based on the diversity, the

assemblage and regularity according to the vintage year.

In recent years the eco physiological approach of vine-environment interaction, as assessed by

the phenotypic expression of the production and quality, has led to a better assessment of the

16

factors that contribute to the wine terroirs. Many of the investigations conducted under

conditions comparable of vineyard model and management system were aimed at the

identification of vocational units (UV) characterized by vegetative performance, production

and quality consistently homogeneous.

The variety „Sangiovese‟ is a genotype characterized by a wide variation of expression due to

its high responsiveness to the environment (Egger et al., 1999; Bandinelli et al., 2001;

Bertuccioli et al., 2001; Giannetti et al., 2001; Scalabrelli et al., 2006), so it would be possible

to obtain in different areas wines with very similar quality levels , though differentiated , thus

expressing the varietal potential in response to a specific terroir (Brancadoro et al., 2006).

1.2.2 The vineyard factors

The „Sangiovese‟ with over 67.4% of the vineyard surface cultivated represents the main

variety for wine production in Tuscany (ARTEA, 2008).

Without a doubt the wines produced with this variety (pure or base for blend) are famous

above all for their geographic origin, thanks to the influence of the environment that

modulates their characteristics (Brancadoro et al., 2006; Storchi et al., 2006). Several Tuscan

Denominations of Origin (DOCG) thanks to their peculiar aspects are distinguished and have

acquired reputation and notoriety abroad (tab. 1), according to several strategies of wine

valorisation (Cotarella, 2001; Gallenti and Cosmina, 2001). Only „Brunello di Montalcino‟

and few other famous red wines are produced with 100% of „Sangiovese‟ while many other

red wines, including those reporting the indication „Sangiovese‟, are produced mainly by this

grape which is integrated with other local or international varieties (Boselli, 2006; Fregoni,

2006; Zampi, 2006). This choice depends on several reasons, to note, problems of inconstant

quality levels to produce aging wines due to insufficient content of anthocyanins and

polyphenols or aromatic pattern, which could occur in some vintage years or in territory with

fair vocation.

17

Denomination Main variety

Number

of

Inscription

Vineyard

area

(hectare)

Maximum

yield of grape

(Ton/hectare)

BOLGHERI

Cabernet

Sauvignon,

Cabernet franc,

Merlot, Sangiovese,

Sirah 263 1.927,18 10

BRUNELLO DI

MONTALCINO Sangiovese 311 2.020,11 8

CARMIGNANO Sangiovese + other

varieties 34 215,72 10

CHIANTI Sangiovese 6160 23.585,45 9

CHIANTI CLASSICO Sangiovese 1178 7.559,14 7,5

MONTECUCCO

SANGIOVESE Sangiovese 317 687,14 9

MORELLINO DI

SCANSANO Sangiovese 412 1.543,54 9

SUVERETO

Sangiovese,

Cabernet

Sauvignon, Merlot, 100 239,03 9

VERNACCIA DI SAN

GIMIGNANO

Vernaccia di San

Gimignano 180 777,88 9

VINO NOBILE DI

MONTEPULCIANO Sangiovese 318 1.286,18 8

Table 1. Wines of excellence produced in Tuscany (DOCG in 2011).

The vineyard model (rootstock, density of plantation, training and pruning system, and type of

management) is of great importance on vine production equilibrium and therefore on the

vineyard ecosystem functioning. Although the „Sangiovese‟ grapevine was well studied, it is

not possible to indicate a generalized vineyard model suited to all situations (Intrieri, 1995;

Loreti and Scalabrelli, 2007). Holding into account the vineyards renewal initiated from 1990'

with the objective to avoid vine vigour excesses, in the new vineyards established in Tuscany

less vigorous rootstocks and close distances of plantation were used to achieve root

competition and decrease vine vigour, which unfortunately not always reached the desired

goal. The rootstocks can offer interesting opportunities for modulating vegetative, yield and

quality performances of vines, especially in difficult situations. In presence of not limiting

conditions less vigorous rootstocks like 3309C and 101-14 can be used, while it would be

more appropriate to use 110R when there is the risk of summer water deficit (Di Collalto et

al., 2001; Scalabrelli et al., 2001; 2003; Palliotti et al., 2006).

During the last phase of vineyards renewal in Tuscany the dominant tendency has been to

adopt planting models with middle or high density planting to achieve better light interception

18

and improve the vine efficiency. This choice is nearly always a result of a compromise

between the best solution from the physiological point of view and the management costs,

which can vary according to the conditions of the territory, the size of the farm and the type of

enterprise. In this region the tendency to adopt a vertical canopy with horizontal cordon spur

pruning or Guyot (less used) is quite generalized, while other innovative systems are

introduced only on a small scale (Intrieri and Poni, 2000; Loreti and Scalabrelli). The optimal

plantation density from the physiological and qualitative point of view for the most part are

intermediate (5000 - 6000 vines/ha), while only in poor soil, is it possible to plant vines at

closer spacing (Scalabrelli et al., 2001; Bertuccioli et al., 2001; Bagnoli et al., 2001; Loreti at.

al., Mattii et al., 2005). Narrow spacing between lines often induces canopy height to decrease

too much with negative effects on grape quality (Scalabrelli et al., 2006). Moreover the model

of vineyard must always be adequate to the environment and technical conditions so as to

obtain a high grape quality potential. Hence much more the behaviour of the vineyard is

approached to the optimal one, the less management actions to equilibrate the system will be

required (fig. 6).

Figure 6. Balance of factors in vineyard ecosystem (Scalabrelli, 2013).

19

1.2.3 Terroir of Tuscany

The spatial soil and climate variability of the different appellation of origin contributes to the

different expression of „Sangiovese‟ has been highlighted by a series of works of

characterization and zoning .This has led to the definition of specific territorial units (UT) and

the drafting of maps of soil landscape to different scales within which a large number of

vineyards were researched in detail (Brancadoro et al., 2006). The climatic survey,

fundamental for the knowledge of terroir, proposes to draw up a climatic map organised into

levels of the territory space, in order to find microclimates of the vineyard, the local climate

and the regional climate (Zamboni, 2003; Carbonneau, 2003; Orlandini et al., 2003). The use

of simple and complex variables or climatic indices, useful for identifying homogeneous areas

requires well distributed information systems and weather stations on the territory and the

adoption of a different scale of reference (Vaudour, 2005; Costantini et al., 2006). The work

conducted in Tuscany by ARSIA meteorological service does not highlight uniform

conditions between the various areas with the possibility of a strong influence of the variable

climate the behaviour of „Sangiovese‟ performances (Scalabrelli, 2008).

The geologic and climatic survey has proved crucial in zoning studies to identify the UTB and

the soil map. In these studies the selected data for physical and chemical analysis of the soil,

the morphological interpretation of horizons, the roots profiles, the micro morphological

analysis and the study of the water functioning or heat of the soil (soil maps with their

properties) were separated from the spatial data (Costantini et al., 2006).

The experiences carried out on various scales in the territory of the provinces of Florence and

Siena showed great differences in the territory regarding to the climate and soil type profiles

and in the behaviour of the productive performance of „Sangiovese‟ variety in several

vineyards included in the main zones of Denomination of Origin: „Chianti‟, „Chianti

Classico‟, „Chianti Colli Senesi‟, „Chianti Colli Fiorentini‟, „Orcia‟, „Nobile di

Montepulciano‟ and „Brunello di Montalcino‟ (Campostrini e Costantini, 1996; Costantini et

al., 1996; Bogoni, 1998; Cricco e Toninato, 2004; Storchi et al., 2005). Climatic variables and

altitude were useful in characterizing only partially the zones having viticulture vocation,

while the mean soil temperature well characterized soils of the „Montalcino‟ vineyards. From

the geologic point of view the viticulture areas of „Montepulciano‟ and „Colli senesi‟ (Siena

hills) proved to be more homogenous, followed by those of Montalcino, while the „Chianti

Classico‟ and the „Orcia‟ resulted much more variable (Costantini et al., 2008). The soils of

„Montalcino‟ and in the „Chianti Classico‟ area were found stoniest and less deep, the latter

20

were also sandiest and could lesser withhold water although exhibiting the best inner water-

drainage, which is considered a key soil factor for the health of the vine root system

(Costantini et al., 2006; Costantini e Bucelli, 2008).

Other soil features like the cationic exchange capacity, the apparent density and the stability

of structure, were found to be quite different in the examined zones. The best vine responses

were observed in selected sites of „Montalcino‟, compared to „Chianti Classico‟. In several

viticulture zones in Tuscany, having the same meteorological conditions, we can find soils

with different physical and chemical characteristics (texture, water-drainage, water content)

that can characterize various units of soil landscape (USL). Examples are furnished by

„Cerreto Guidi‟ (Cricco and Toninato, 2000), and „Bolgheri‟ (Bogoni, 1999) where the

expression of „Sangiovese‟ is identified by different wine sensory profiles.

In a study carried out in the „Arezzo‟ province, 33 vineyards of `Sangiovese' were subdivided

into groups of premature sugar accumulation underlining the greater importance of the inner

soil water-drainage (26%), followed by altitude (16%) and water availability (11%) (AWC =

Available Water Content). The factors water-drainage, AWC and altitude well discriminate

the performances of the vineyards, important too are the depth of soil and the texture (Fig.

11). In particular the water-drainage and the AWC influenced significantly the kinetic of

sugar accumulation and the main qualitative grape parameters at harvest. The early ripening

vineyards achieved the best grape quality (higher anthocyanins and polyphenols content) and

also gave wines more interesting sensory profiles and of greater amplitude. Surveying has

permitted a subdivision of the territory into territorial units (UT) having similar characteristics

of soil, landscape and climatic conditions, and productive and qualitative expressions

(Toninato et al. 2005).

Several soils identified in „Montalcino‟ area having different water available in the ground

during ripening proved to significantly influence most characteristics (sugars, pH and acidity)

according to the year, while the extractable anthocyanins and polyphenols content were

influenced only by vintage year. It appeared obvious that in order to obtain the best

characteristics from `Sangiovese, soil water content is crucial during the period between

veraison to ripening, during which a moderated water deficiency is positive, while excesses or

drastic reductions of water available are both negative (Storchi et al., 2000).

Work conducted on small scale zoning in „Montalcino‟ area allowed to identify territorial

units (UT) characterized by soil chemical and physical parameters (table 3) eg. texture and

electrical conductivity, an indirect evaluation of AWC. In this case the UT richer in clay was

able to assure the greater AWC in the ground during maturation, while the other UT induced

21

in `Sangiovese' conditions of water stress from severe to light. Such conditions have

influenced meaningfully the kinetic of ripening and particularly increasing the sugar

accumulation and the content of extractable anthocyanins (Brancadoro et al., 2006). These

results are in agreement with several papers which have underlined that a moderated water

stress occurring in the period between veraison and ripening, induces favourable

characteristics of the grapes and the wine (Scalabrelli, 2006; Remorini et al., 2007).

Extensive research activities performed in the province of Siena have confirmed that the vine

root depth and AWC are significantly correlated with viticulture parameters while the vine

performances were mainly influenced by the annual variables like the temperatures of the

ground and the air, the rainfall and the number of days without rain (Costantini et al., l.c.). At

the end of a series of surveys led in this province an innovative methodology for the terroir

definition was proposed which previewed the construction of a soil information GIS which

collected data in geographic form (scale 1:100.000) and alphanumeric (DB) data. Moreover a

DB of viticulture and oenological data, monitored in 70 vineyards for several years, was

created. On the basis of these results in this province 363 terroir were thus characterized that

have an average extension of 46 hectares, varying from a minimum of 2 to a maximum of 474

hectares (Costantini et al., l.c.). Therefore, it can be noted how in Tuscany there are many

viticulture terroir although very few are really homogenous. The majority of the terroir can

be considered multiple sites, in which the meaningful effect of the soil and climatic factors,

utilized to characterize the UV, on the qualitative characteristics of the grapes and wines, can

determine a compensatory effect in areas of non homogenous production supplying however

wines having similar characteristics (Brancadoro et al. l.c.).

Recent studies on grapes were able to predict wine characteristics (Bucelli et al., 2010), while

berry sensorial analysis was used as a complementary method to determine berry quality

(Ducci et al., 2012) and the study of the aromatic profile as well (D‟Onofrio et al., 2012).

22

1.3 Aim of the thesis


its high responsiveness to the environment so it is possible to obtain in different areas wines

with quality standards very similar, though differentiated between them, thus expressing the

varietal potential in response to a specific terroir. Although the „Sangiovese‟ grapevine was

well studied, it is not possible to indicate a generalized vineyard model adapt to all situations.

The project, through a structured study wants to proceed first to the chemical-physical and

sensory characterization of „Sangiovese‟ grapes produced in some representative production‟s

areas of Tuscany. Through a detailed study on the main components responsible for the quality

of grapes, especially aroma compounds, based on ecopedological factors, it will deepen the

knowledge of the factors that in the vineyard are the source of differentiation in order to

provide the necessary tools to operate more efficiently technical choices that can make a

valuable contribution to the diversification and the identification of the wines produced. These

investigations, does not aim to make a hierarchical scale of oenological products of a specific

territory, but to provide a way to understand the potential of a territory. In this work a particular

attention was dedicated to the areas of „Montecucco‟ and „Brunello di Montalcino‟ making a

focus on vineyard effect („ColleMassari‟ estate) and on clone effect („Col d'Orcia‟ estate).

23

2. MATERIALS AND METHODS

2.1 Plant material

The research was conducted in three consecutive years (2009, 2010 and 2011), on

„Sangiovese‟ vineyards in five areas of production located in „Grosseto‟, „Pisa‟, and „Siena‟

provinces, involving a total of 17 theses (fig. 7 and tab. 2).

The corresponding Denomination areas of wine production were: „Brunello di Montalcino‟,

„Chianti Classico‟, „Chianti Colline Pisane‟, „Montecucco‟ and „Morellino di Scansano‟. On

„Montecucco‟ and „Brunello di Montalcino‟ the study were focused on several vineyards and

the clonal effect was also studied. The vineyards in our study were planted in 2000-2001, at a

density of 4000-5000 plants per hectare, trained to horizontal spur pruned cordon. The use of

cloned material has almost always been identified, most of these clones are „R24‟ and „SS-F9-

A548‟ grafted on 420A or 1103P rootstock depending on the cultivation area (tab. 3).

Figure 7. Main D.O. and D.O.C.G. areas in Tuscany on Sangiovese vines.

Carmignano

Chianti e Chianti Classico

Brunello di Montalcino

Sovana

Nobile di Montepulciano

Carmignano

Chianti e Chianti Classico

Brunello di Montalcino

Sovana

Nobile di Montepulciano

Capalbio

Morellino di Scansano

Monteregio di Massa Marittima

Val di Cornia

Colline Lucchesi

Montescudaio

Capalbio

Morellino di Scansano

Monteregio di Massa Marittima

Val di Cornia

Colline Lucchesi

Montescudaio

Montecucco

Chianti Colline Pisane

24

Table 2. Prospect of vineyards chosen for the study.

Table 3. Main characteristics of the vineyards (SPC = Spur pruned cordon; C= Current; B=Biologic;

BD= Biodinamic.).

Thesis Code Company - Vineyards Denomination Area Town Prov. Site

1 CCP 1 Beconcini Chianti Colline Pisane S. Miniato Pi 1

2 MC 1

ColleMassari

Campo La Mora F9 Montecucco Cinigiano Gr 2

3 MC 2

ColleMassari

Campo La Mora Sal “ Cinigiano Gr 2

4 MC 3 ColleMassari Cerrete “ Cinigiano Gr 2

5 MC 4 ColleMassari Orto del Prete “ Cinigiano Gr 2

6 MC 5 ColleMassari Vigna Vecchia “ Cinigiano Gr 2

7 MC 6 Salustri “ Cinigiano Gr 3

8 MS 1 Fattoria di Magliano Morellino Scansano Magliano Gr 4

9 BM 1 Col D'Orcia

Brunello di

Montalcino Montalcino Si 5

10 BM 2 Col D'Orcia “ Montalcino Si 5




14 BM 6 Casanova Di Neri “ Montalcino Si 5

15 BM 7 La Mannella Terra Bianca “ Montalcino Si 6

16 CC 1 Capannelle Chianti Classico Gaiole Si 7

17 CC 2 Castello di Albola Chianti Classico Radda Si 8

Thesis Code Clone

Rootstock

Training

system Age Conduct

1 CCP 1 F9 1103P SPC 10 C

2 MC 1 F9 161-49 C Guyot 9 B

3 MC 2 Sel. Salustri 161-49 C Guyot 9 B

4 MC 3 Sel. Salustri 110 R Guyot 8 B

5 MC 4 Sel. Talenti 157-11 SPC 9 B

6 MC 5 Sel Col D‟Orcia 775 P SPC 10 B

7 MC 6 Sel. Salustri 110 R Guyot 10 B

8 MS 1 R 24 110 R SPC 10 C

9 BM 1 Clone 1 420A SPC 10 C





14 BM 6 VCR 5 110 R SPC 9 C

15 BM 7 R 24 1103P SPC 10 C

16 CC 1 R 24 420A SPC 10 C

17 CC 2 R 24 420A SPC 9 C

25

2.2 Technological maturity

At harvest time, sets of 10-12 bunches for thesis were sampled and subjected to sensorial,

physical-chemical, and aromatic analyses. Crashed bunches were used to determine the

concentration of total soluble solids (°Brix) by a digital refractometer (Model 53011, TR,

Forlì, Italy), the pH by a bench pH-meter (Hanna Instruments, Milano, Italy) and total acidity

by a digital burette (Brand, Wertheim ,Germany) by titration with NaOH 0.1 N.

2.3 Sensory analysis on grapes

Evaluating the quality of the grapes before harvest is very important for the decision making

of the vine technician and the enologist, as they can better direct the growing techniques in

vineyard and better select the grapes for the wine making based on quality and suitability in

the production of specific wines. Moreover it is possible to adjust the winemaking technology

on the basis of the characteristics of the raw material. The ICV has in the last ten years

developed a sensorial method of analysis in order to satisfy the above needs (www.icv.fr).

This method, slightly modified (Scalabrelli et al., 2010) can be used with good results at

contained costs, above all because it represents a complementary technique to the chemical-

physical analysis of the grapes (sugar, acidity, phenolic ripeness).

Sensorial analysis of the grapes consists in the evaluation of the visual and tactile

characteristics of the berry by the sequential tasting of the skin, the pulp and seeds.

The method used obtains the evaluation by one test: a) the mechanical characteristics of the

single berry, acid balance, aromatic strength, quantity and quality of polyphenol and the

respective localization; possible imbalance in ripeness levels of the different parts of the

berry; c) the variation of the technological ripeness in different periods and years. The

procedure expects that every wine taster on the panel assigns for every single descriptor a

mark from 1 to 4 corresponding to a level of increasing ripeness (tab. 4-5).

However it must be noted that for some parameters higher values correspond to an advanced

level of berry ripeness. This is the reason for the astringency of the tannin and the sensation of

bitterness.

Such requirements from a technical point of view are very important in the life of the vine and

in the conditions in which ripening occurs, as out of phase ripening can be important from a

technological point. Therefore from a sensorial analysis by a trained panel of tasters, it is

possible to understand the differences in the stage of ripeness of the different parts of the

berry, that can be difficult to determine analytically. For example the sensation of astringency

26

bitterness and aroma can be quantified immediately without turning to laboratory analysis, but

for sweetness and acidity of the must laboratory analysis is needed.

The panel of tasters were five and they were specifically trained in the tasting procedure.

Part/Score Sensorial descriptors

Berry Skin color Plasticity Pedicel

detachment

1 Pink Hard Very difficult

2 Red Elastic Difficult

3 Red dark Plastic Easy

4 Blue - black Easy to break Very easy

Skin Aptitude to the

Skin grinding

Tannins

astringency

Aromatic

dominant notes

Bitter

sensation

1 Very difficult Strong Herbaceous Strong

2 Difficult Medium Neutral Medium

3 Little hard Light Fruity Light

4 Tender Nothing Marmalade Little

Pulp Flesh - Skin

separation

Sweet

sensation Acid sensation

Aromatic

dominant notes

1 Very tight Little High Herbaceous

2 Middle tight Medium Medium Neutral

3 Tight Sweet Little Fruity

4 Not tight Very sweet Nothing Marmalade

Seed Colour Hardness Tannins

astringency

Aromatic

dominant notes

Bitter

sensation

1 Green Soft High Herbaceous Strong

2 Green -brown Little soft Medium Neutral Medium

3 Brown Hard Little Little roasted Light

4 Brown dark Lignified Nothing Roasted Little

Table 4. Synthetic sheet of sensorial descriptors and relative score attributed to each level perceived

during tasting of the berry parts. (Method proposed by Department of Fruit Science and Plant

Protection of Woody Species: Scalabrelli, 2008).

Score Skin color Skin Flesh Grape-seed Technological

evaluation

1 Little colored

Tight to pulp,

herbaceous taste,

astringent

Hard, acid, and

herbaceous

taste

Green, gummy,

astringent and

bitter

Unripe

2

Incomplete

coloration with

green veins

Tight to pulp,

lightly

herbaceous, thick,

astringent

Thick, acid,

herbaceous

taste

Brown with green

veins partially

lignified astringent

and bitter

In progress of

ripening: not to

harvest

3 Red, enough

uniform

Thick, a little bit

fragile, lightly

fruity with final

herbaceous taste

Little thick,

lightly acid

Brown, lignified

and little

astringent

Almost ripe, to

vinification

with particular

attention

4 Bleu - Black Easy to remove,

fragile, strong

Sweet juicy

flesh, fruity

Lignified, spiced

or lightly hot,

Ripe: suitable

to make wine

27

fruity notes

without final

herbaceous taste

and/or

marmalade

taste

almond taste, not

bitter and not

astringent.

Table 5. Sensorial analysis of berry parts (with black skin) during ripening: score and synthetic

description of perceived sensations.

2.4 Phenolic compounds analysis

For each sampling, 60 berries were randomly chosen, divided into three groups of 20 berries,

which were used as triplicates, and processed according to the method of Di Stefano et al.

(2008) slightly modified as follows. Berry skins of each replicate were manually separated

from pulp and seeds, and skins and seed were separately weighed and extracted for 4 h at

25°C in 25 mL of a pH 3.2 tartaric buffer solution. This solution contained 12% (v/v) ethanol,

2 g/L sodium metabisulphite, 5 g/L tartaric acid and 22 mL/L NaOH 1 N. After grounding in

a mortar and pestle, the extract was separated by centrifugation (R-9M: Remi Motors TD,

Vasai India) for 10 min at 3000 rpm. The pellet was re-suspended in 20 mL of buffer and

centrifuged for 5 min. The final two pooled supernatants were adjusted to 50 mL with the

buffer solution. The skins extract was measured by UV-Vis absorption (Spectrophotometer

HITACHI U-2000) at 540 nm after dilution (1:20) with ethanol: water : HCl (70:30:1) and at

750 nm as the seeds extract in the following solution: 0.1 mL of the extract, 6 mL H2O, 1 mL

Folin-Ciocalteu reactive, 4 mL 10% Sodium Carbonate (after 5 min) and H2O up to 20 mL.

Anthocyanins were expressed as mg of equivalents of malvidin 3-O-glucoside and phenolic

compounds as mg of equivalents of (+)-catechin.

2.5 Aroma compounds analysis

Aroma compounds originating from the enzymatic hydrolysis of glycosidic precursors

(aldehydes, benzene derivates, monoterpenes, norisoprenoids) were extracted from fresh

berries by Solid Phase Extraction (SPE) according to the protocol described by Di Stefano et

al. (1998).

Moreover to reproduce changes in compounds occurring during ageing, hydrolysis of the

extract was performed under similar acidic conditions of wines. In order to have a total

overview and concentration of aroma compounds, we decided to carry out hydrolysis reaction

on the methanolic extract obtained after enzymatic hydrolysis.

28

Preliminary investigations, considering the use of SPE or SPME procedure for the separation

and analysis of the formed compounds, showed SPME technique more suitable because of its

efficiency in the extraction, taking into account also the very low concentration of the

revealed compounds.

Free compounds were not considered in this work because usually their contribution in

neutral grapes, as „Sangiovese‟, is very limited.

2.5.1 Preparation of grape sample

Skins of 100 berries were separated from the pulp and extracted with 20 mL of methanol for 1

hour while the pulps were put in a glass containing 100 mg of sodium metabisulfite. After 1

hour pulp and juice were reunited with 150 mL of a pH 3.2 tartaric buffer solution (2 g/L

sodium metabisulphite, 5 g/L tartaric acid and 22 mL/L NaOH 1 N). After homogenization by

Ultra - Turrax and centrifugation at 7000 g for 5 minutes, solid parts were washed with 100

mL of pH 3.2 tartaric buffer solution and again centrifuged, and the clear liquid was reunited

to the first one. The obtained extract was treated with pectolytic enzyme (Vinozym FCEG) for

1 night at room temperature and finally filtered (Whatman 42).

2.5.2 Enzymatic hydrolysis of glycosides

The extract was added to 200 μL of 1-heptanol (40 mg/L in ethanol) as internal standard, and

the solution was passed through a cartridge 5 g C18 Sep Pak (WAT 036795) previously

activated by 20 mL methanol, and 50 mL water. After the sample loading, salts, sugars, and

more polar compounds were removed by washing the cartridge with 100 mL of water, and the

fraction containing free compounds was recovered by elution with 30 mL of dichloromethane.

A second fraction containing glycoside compounds was recovered with 30 mL of methanol.

The methanolic solution was evaporated to dryness under vacuum at 40 ° C, the residue was

dissolved in 5 mL of a citrate – phosphate buffer pH 5 (2.04 g of citric acid, 2.92 g of

hydrogen phosphate monoacid ), then it was added to 200 μL of a glycosidic enzyme with

strong glycosidase activity and kept at 40 °C overnight. Then, the solution was centrifuged,

added to 200 μL of a 1-heptanol (40 mg/L), and the resulting solution was passed through a 1

g cartridge C18 Sep Pak (WAT 036795) cartridge previously activated by 5 mL methanol and

10 mL water. After cartridge washing with 10 mL of water, the fraction containing the

aglycones was eluted with 6 mL of dichloromethane, dehydrated with sodium sulfate

anhydrous, and concentrated to 200 μL before analysis. A last fraction, containing the

29

potentially aromatic precursor compounds, was recovered from the cartridge by elution with 5

mL methanol.

2.5.3. Acid hydrolysis of glycosides

The methanolic solution, was evaporated to dryness under vacuum at 40 ° C and the residue

was dissolved in 10 mL of tartrate buffer at pH 3.

1g of sodium chloride and 8μL of a 40 mg/L solution of 1-heptanol were added to this

solution and the mixture was heated in a water bath to 100°C for 1 h, in an encapsulated vial

under a nitrogen atmosphere. After cooling, 2.5 mL of the resulting reaction mixture was

transferred in a 20 ml headspace vial and extracted with a SPME fiber (DVB/CAR/PDMS)

using an automatic CombiPal system (CTC analytics) under the following conditions:

incubation at 60°C for 20 min.; extraction for 35 min.; desorption in the GC injector at 240°C

for 6 min in pulsed splitless mode (25 psi for 5 min).

2.5.4. Gas chromatography – mass spectrometry

Chromatographic analysis were carried out using a Agilent 7890A gas-chromatograph

coupled with a Agilent 5975C quadrupole mass spectrometer. The carrier gas was helium at a

constant flow rate of 1 mL/min. The capillary column was a HP-Innowax (30 m length, 0.25

mm i.d., 0.25 mm film thickness) from Agilent. The temperature programme of the column

oven started at 30 °C, then increased at 30 °C/min to 60 °C for 2 min, at 2 °C/min to 190 °C,

and at 5 °C/min to 230 °C for 10 min. The MS detector scanned within a mass range of m/z

30-450.

Compounds were identified by a combination of matching retention indices with library

matches (Nist 08) and authentic standards, which were available for the compounds of

interest. The quantification was carried out comparing the peak area of each compound with

that of the internal standard.

30

2.6 Statistical analysis

The resulting data was then analyzed statistically using SPSS130 software. In detail, the

climatic data underwent cluster analysis and discriminating analysis and the visual results by

centroids which report the first two canonical functions. The macro and microstructural

characteristics data of the grapes at harvest were analyzed using the MANOVA test and the

differences highlighted two by two by the Tukey test.

With the statistical analysis by placing the three factors thesis, year, and the interaction thesis

by year, it was calculated the percentage of variance due to each factor relative to the total

variance, obviously including the error.

A factorial statistic analysis was also carried out to reduce the wide variability range of

descriptors and constitute complex variables that could well represent the theories examined

and to highlight the substantial differences that exist among them. Subsequently the results

underwent multiple linear regression to show the possible correlation between the parameters

examined.

The evaluation given in the berry sensorial analysis were transformed in percentages before

statistical analysis.

Statistical significance was accepted at P <0,05.

31

3. RESULTS

3.1 Weather station location and study of the historical sequences

The ARSIA archive provided the data relative to the areas under study by selecting six

weather huts that were close to each other and that were representative of the production area.

The climatic characteristics of the areas were analyzed by studying the following parameters:

min-max temperatures, temperature range and rainfall. Average values were calculated for the

period April–October and the bioclimatic index Growing Degree Days using the total daily

temperatures >10°C. As regards the station chosen on a specific site corresponding to n°2

(„ColleMassari‟, „Montecucco‟), the climatic findings were collected from weather huts

installed on the company.

The localization of the weather huts (fig. 8) and the climatic characteristics of the different

areas of Tuscany can be seen in fig. 9-11 relative to average annual temperatures, annual

rainfall (mm) and to the hydroclimatic balance (mm) up to the year 2007 (the difference

between the total rainfall and the total ETP).

Figure 8. Weather stations used (initials corresponding to table 6).

3

2

1

4

5

6

32

The data collected is in tabulate and graphic format so as to compare the different areas (tab. 6

a,b,c,d).

n Station

Temp

max

(°C)

Temp

min

(°C)

Temp

media

(°C)

Daily excursion

(°C)

Rainfall

(mm)

Growing

Degree

Days (°C)

Huglin

Index

1 San Miniato 26,2 14,7 21,5 11,5 234 2.043 2915,4 2 Cinigiano* 25,6 14,6 20,6 11 254 1.943 2548,8 3 Magliano 26,8 16,3 21,8 10,5 275 2.159 3206,8 4 Montalcino 23,8 14,3 19,5 9,5 331 1.750 2315,8 5 Montalcino** 25,6 13,7 20,3 11,9 254 1.895 2776,9

6 Gaiole 26,0 9,5 18,7 16,5 292 1.608 2449,8 * Poggi del Sasso

** Argiano

Table 6 (a). Average data about weather trends in the year 2009 in the period April-October.

n Station

Temp

max

(°C)

Temp

min

(°C)

Temp

media

(°C)

Daily excursion

(°C)

Rainfall

(mm)

Growing

Degree

Days (°C)

Huglin

Index

1 San Miniato 24,1 13,0 19,2 11,1 450 1.688 2470,3 2 Cinigiano* 24,2 13,7 19,3 10,5 379 1.715 2243,8 3 Magliano 23,1 14,2 19,4 8,9 355 1.739 2991,8 4 Montalcino 22,4 13,2 18,2 9,2 449 1.509 1981,8 5 Montalcino** 23,7 12,7 19,0 11 442 1.732 2374,6 6 Gaiole 24,2 9,0 17,4 15,2 465 1.368 2081,9

* Poggi del Sasso

** Argiano Table 6 (b). Average data about weather trends in the year 2010 in the period April-October.

n Station

Temp

max

(°C)

Temp

min

(°C)

Temp

media

(°C)

Daily excursion

(°C)

Rainfall

(mm)

Growing

Degree

Days (°C)

Huglin

Index

1 San Miniato 26,5 14,2 20,9 12,3 185 2.181 2902,2 2 Cinigiano* 25,6 14,3 19,5 11,3 344 2.051 2623,1 3 Magliano 27,1 15,1 21,8 12 227 2.791 3389,6 4 Montalcino 24,1 14,2 19,3 9,9 349 1.847 2389,6 5 Montalcino** 25,6 13,2 20,6 12,4 352 2.138 2753,3 6 Gaiole 26,4 9,3 18,7 17,1 249 1.662 2599,6

* Poggi del Sasso

** Argiano Table 6 (c). Average data about weather trends in the year 2011 in the period April-October.

33

n Stazione

Temp

max

(°C)

Temp

min

(°C)

Temp

media

(°C)

Daily excursion

(°C)

Rainfall

(mm)

Growing

Degree

Days (°C)

Huglin

Index

1 San Miniato 25,6 14,0 20,5 11,6 290 1.971 2762,6

2 Cinigiano* 25,1 14,2 19,8 10,9 326 1.903 2471,9

3 Magliano 25,7 15,2 21,0 10,5 286 2.230 3196,1

4 Montalcino 23,4 13,9 19,0 9,5 376 1.702 2229,1

5 Montalcino** 25,0 13,2 20,0 11,8 349 1.922 2634,9

6 Gaiole 25,5 9,3 18,3 16,2 335 1.546 2377,1 * Poggi del Sasso

** Argiano Table 6 (d). Average data relating to meteorological trends of 2009-2011 three-year period during

the reference period April-October.

Figure 9. Territorial distribution of the average annual temperature (period 1998-2007).

ARSIA source.

3

2

1

4

5

6

34

Figure 10. Territorial distribution of annual rainfall (mm/per year) (period 1998-2007).

ARSIA source.

Figure 11. Regional distribution of the hydroclimatic balance (mm) for the year 2007 (the difference

between the total rainfall and the total ETP). ARSIA source.

3

2

1

4

5

6

3

2

1

4

5

6

35

3.2 Climatic characteristics

The graphic representation of some averages highlight the differences between areas relative

to average temperatures, temperature range, rainfall and total of Growing Degree Days. (figures

12-23).

The „Gaiole‟ station shows in all the three years the lowest average temperature rates, while

the province of „Grosseto‟, with „Magliano‟ in 2009 and 2011 and with „Cinigiano‟ in 2010

reached the highest average temperatures in the period April-October. In addition „Gaiole‟

differs greatly from the other stations for reaching the highest temperature range in the three

year period, with temperatures close to 20°C in the main summer months. Growing Degree

Days have shown different trends for the years 2009-2010 where there was an accumulation

up to the month of August followed by a sharp fall; instead in 2011, the GDD do not all

follow the same trend moving away from the values of the previous years. To note that 2010

represents the year with the highest rainfall in mm, with the exception of „Argiano‟, where the

mm rainfall is lower compared to the other two years studied.

Figure 12. Average temperature trend in 2009 during the reference period April-October.

36



37

Figure 15. Daily excursion trend in 2009 during the reference period April-October.



38

Figure 18. GDD trend in 2009 during the reference period April-October.



39

Figure 21. Rainfall trend in 2009 during the reference period April–October.



40

By cluster analysis of the variables noted it was possible to obtain a dendrogram where it was

possible to see areas with similar climatic conditions on the basis of the mean climatic data

observed in each year of the three years studied from the six stations (fig. 24). The sites in the

study are grouped in cluster, the first being „Col D‟Orcia‟ 2010 („Argiano‟) and the second all

the other weather stations. In the second group only the „Magliano‟ 2011 station is

distinguishable from the others. The „Gaiole‟ and „Montalcino‟ stations, unlike the others, fall

in the same group for all three years.

From the centroids graph obtained from the cluster analysis (fig. 25), which highlighted a

significant result since the first two accepted functions represent 93,1% of the total variability,

it is noted that the points relative to the groups show a limited dispersion with the exception

of the „Cinigiano‟ station. Three distinct groups appear in the centroid, the first the stations of

„San Miniato‟ and „Magliano‟ that appear less distinct and the two belonging to the

„Montalcino‟ area; the second the station of „Gaiole‟ very close to that of „San Miniato‟ and

the third „Cinigiano‟ which is the most distinguishable.

The 91,3% of the original grouping are classified correctly, while 87,3% of the cases grouped

cross-validated are reclassified correctly (tab. 8). From the test table of the effects among

subjects, obtained from the multifactor analysis of the variables examined, differences

between weather stations emerge (tab. 7). Analysing the station as a source, the dependent

variables statistically different are maximum/minimum temperature and temperature range. If

on the other hand it is the year as the source the temperature range remains statistically

different together with the rainfall. From the interaction year by station however, the statistics

on the parameters examined resulted negative, that is to say, there are no statistically

significant differences between the stations studied.

41

Figure 24. Dendrogram obtained from the hierarchical cluster analysis of the mean data of the six

stations. Stations grouped on the basis of the average link between groups.

.

Rescaled Distance Cluster Combine

Figure 25. Centroids obtained from the cluster analysis of the climatic parameters.

42

Factor

Dependent

Variable F Sig.

Station T Max 5,717 ,000

T Min 9,235 ,000

T Med 1,615 ,162

Rainfall ,631 ,677

Excursion 58,455 ,000

GDD 1,783 ,122

Year T Max 2,759 ,068

T Min 1,384 ,255

T Med 2,135 ,123

Rainfall 6,359 ,002

Excursion 9,577 ,000

GDD 2,501 ,087

Station

* Year

T Max ,048 1,000

T Min ,058 1,000

T Med ,048 1,000

Rainfall ,262 ,988

Excursion ,260 ,988

GDD ,174 ,998

Table 7. Test of the effects between subjects (p=<0,05).

Station

Number

Group expected

Totals 1 2 3 4 5 6

Cross-

validation

% 1 66,7 ,0 33,3 ,0 ,0 ,0 100,0

2 ,0 95,2 ,0 ,0 4,8 ,0 100,0

3 23,8 ,0 76,2 ,0 ,0 ,0 100,0

4 ,0 ,0 4,8 95,2 ,0 ,0 100,0

5 ,0 ,0 9,5 ,0 90,5 ,0 100,0

6 ,0 ,0 ,0 ,0 ,0 100,0 100,0

Table 8. Classification results.

a. Cross validation is done only for those cases in the analysis in cross validation, each case is

classified by the functions derived from all cases other than that case. b. 93,7% of original grouped cases correctly classified.

c. 87,3% of cross-validated grouped cases correctly classified.

43

3.3 Vineyards’s characteristics

The main characteristics of the vineyards are reported as followed according to the denomination.

3.3.1 ‘Chianti Colline Pisane’

3.3.1.1 ‘Beconcini’ estate

The main component of the soil where the vineyard is situated is the white sand. The other

layers that make up these soils are varied, very thin and mostly consist of a series of marine

fossils from the Pliocene age and of sandstone. So we can see shells of every size and in

quantities such as to constitute, in some cases, the real skeleton of soils and also sands from

the finest to the heaviest, arranged in thin layers and very unlike for salinity and fertility as the

soil goes downwards. The soils are alkaline.

The vineyards are trained to spur pruned cordon with planting design of 1 meter on the row

and 3 meters between rows. With this pruning, every plant presents four spurs with two buds

and so, bud load of eight buds.

Figure 26. „Sangiovese‟s vineyard in „Beconcini‟ estate.

44

3.3.2.‘Montecucco’

3.3.2.1 ‘ColleMassari’ estate

The vineyards of our study are quite homogeneous in terms of conduct, training system,

rootstock and age. Only in the case of „Orto del Prete‟ vineyard, training system is spur

pruned cordon unlike the system used mainly that is the Guyot. All the vineyards have a bud

load of 10 buds per plant.

The main crop operations are performed in the same way in all the vineyards and the

spontaneous grass cover among rows is used.

„ColleMassari‟ is conducted according to the organic protocol and then the pest protection is

carried out exclusively by the use of copper-based and sulphur-based products according to

the limits imposed by law. Shoot thinning is carried out only one time.

The widespread use of summer pruning and mechanic thinning restrict vegetation, which in

itself would be very vigorous. Usually the cluster thinning takes place during the first week of

September.

‘Campo la Mora F9’: sloping vineyard of around 15%, with „rittochino‟ layout in north-east

south-west row orientation and then east- west in the end. The plantation dates from 2003.

Training system: simple Guyot

Clone: clone F9

Rootstock: 161-49 Couderc

Planting design: 2.30 x 0.80 m

Vine density: 5435 vines/ha

Figure 27. „Campo la Mora F9‟ vineyard in „ColleMassari‟ estate.

45

‘Campo la Mora Sal.’: sloping vineyard of around 15%, with „rittochino‟ layout in a north-

east south-west row orientation. The plantation dates from 2003.


Clone: Salustri selection

Rootstock: 161 - 49 Couderc



Figure 28. „Campo la Mora Sal.‟ vineyard in „ColleMassari‟ estate.

‘Cerrete’: sloping vineyard greater than 15%, with „rittochino‟ layout and in north-east

south-west row orientation. The plantation is the youngest among the six examined vineyards:

it planted in 2005


Clone: Salustri selection

Rootstock: 110 R.



46

Figure 29. „Cerrete‟ vineyard in „ColleMassari‟ estate.

‘Orto del Prete’: sloping vineyard greater than 15%, with „rittochino‟ layout and in east- west

row orientation. The plantation dates from 2001.

Training system: spur pruned cordon

Clone: Talenti‟s selection

Rootstock: 157-11 Couderc



Figure 30. „Orto del Prete‟ vineyard in „ColleMassari‟ estate.

47

‘Vigna Vecchia’: sloping vineyard with „rittochino‟ layout and in north - south row

orientation. The plantation dates from 2001 planting out five Sangiovese‟s selections.


Clone: Col D‟Orcia n°5 selectioned clones

Rootstock: 775 P



Figure 31. Vigna Vecchia vineyard in ColleMassari estate.

3.3.2.2 ‘Salustri’ estate

The farm is conducted according to the organic protocol. The ground belongs to those who

present the surface layer from 0 to 30/40 cm, yellowish brown, dry, without any mottling,

with small, common, scarce and medium pores. Besides it is neither very adhesive nor plastic,

with sandy texture, frequent skeleton, devoid of concretions, non-calcareous with pH of 6,8.

The organic fraction of the soil is very low; thus microbial activity, physical structural

characteristics and chemical fertility are adversely affected. The contribution of organic

substance is therefore necessary. The cationic exchange capacity (C.E.C.) is average; the

amount of nutrients kept in cationic form is good. Total nitrogen appears to be low; his

contribution to nitrogenous nutrition of crop is modest. The phosphorus level is medium while

the calcium level is low, as well compared with C.E.C.

The vineyard was established with a „Sangiovese Salustri‟ selection at 0,8 meter on the row

and 2,3 meters between rows (5700 vines/ha), trained to Guyot with 8 buds per plant.

48

Figure 32. Sangiovese‟s vineyard in „Salustri‟ estate.

3.3.3 ‘Morellino di Scansano’

3.3.3.1 ‘Fattoria di Magliano estate’

The soil where the vineyard is situated has a moderately rich texture of skeleton and it is

calcareous. Is situated at 150 m above sea level, in a south-west row orientation.

The pH of the soil is alkaline.

The vineyards in our study are trained to unilateral spur pruned cordon with planting design

of 0,8 meter on the row and 2,2 meters between rows vines are hedged at 0,8 m in height. The

average yield is 1 kg/vine.

Figure 33. Sangiovese‟s vineyard in „Fattoria di Magliano‟ estate.

49

3.3.4 ‘Brunello di Montalcino’

3.3.4.1 ‘Col d’Orcia’ estate

The vineyard is placed on a slight slope with „rittochino‟ layout and in north - south row

orientation. The soil is silty, sandy, alkaline, calcareous reaction with active medium lime,

with a low content of organic substance and of potassium but is characterized by high strength

of magnesium, and of calcium and medium C.E.C. The vineyard was planted in 2000 with

planting design of 0,8 meter on the row and 2,35 meters between rows and 5319 vines/ha is

the planting density.

In this vineyard there are five Sangiovese‟s clones in selection (virus-free) on 420A rootstock,

this ensures a limited vigour, even considering the low amount of organic substance and total

nitrogen.

The vineyard is trained to spur pruned cordon with four spurs per plant; cluster thinning is

used to maintain production within disciplinary levels DOCG „Brunello di Montalcino‟

cluster thinning is used.

Figure 34. Sangiovese‟s vineyard in „Col D‟Orcia‟ estate.

3.3.4.2 ‘La Mannella’ estate

The vineyards is to the north-east of „Montalcino‟, is trained to spur pruned cordon placed at

at 0,8 m from the soil with rows in south east row orientation. The planting design is of 0,8

meter along the row and 3 meters between the rows. The „Sangiovese‟ clone is R24 grafted on

1103P. The soil is stony and alkaline.

50

Figure 35. Sangiovese‟s vineyard in „La Mannella‟ estate.

3.3.4.3 ‘Casanova di Neri’ estate

The vineyard is trained to spur pruned cordon and vines are hedged at 0,55 m in height in

south east row orientation. The planting design is of 0,8 meter on the row and 2,2 meters

between rows. „Sangiovese‟ is a mass selection grafted on 110R. The soil has a medium

consistence and it is rich in stones.

Figure 36. Sangiovese‟s vineyard in „Casanova di Neri‟ estate.

51

3.3.5 ‘Chianti Classico’

3.3.5.1 ‘Castello di Albola’ estate

The vineyard with an east exposure, located in the hills, is placed at an altitude of 550 meters.

The planting design is of 0,80 meter on the row and 2,50 meters between rows and 5000

vines/ha is the planting density.

„Albola‟s soils are characterized by the presence of pedological formations of limestone-

marly nature and partly calcareous-clayey. Morphologically the rock comes from austro-

alpine domain, with a place in the „series of marly limestone‟.

Figure 37. Sangiovese‟s vineyard in „Castello di Albola‟ estate.

3.3.5.2 ‘Capannelle’ estate

The vineyard is collocated on calcareous rocks and it is characterized by rich stones, of

similar origin to the soil of „Albola‟ estate, rows wirh east-west orientation. Vines are trained

to horizontal spur cordon placed at 0,6 m from the ground. At distance of 0,8 x 2,5 m (5000

vine/ha).

Figure 38. Sangiovese‟s vineyard in „Capannelle‟ estate.

52

3.4 Soil’s characteristics

By cluster analysis of the chemical and physical characteristics of the experimental soils (tab.

10) it was possible to obtain dendrograms where it was possible to see theses (tab. 9) with

similar pedologic conditions.

Thesis Code Vineyard Estate

N°

denomination Denomination

1 CCP 1 Beconcini Beconcini 1 Chianti Colline Pisane

2 MC 1 CM F9 ColleMassari 2 Montecucco

3 MC 2 CM Sal ColleMassari 2 Montecucco

4 MC 3 Cer ColleMassari 2 Montecucco

5 MC 4 O_P ColleMassari 2 Montecucco

6 MC 5 V_Vec ColleMassari 2 Montecucco

7 MC 6 S_Marta Salustri 2 Montecucco

8 SC 1 Magliano Magliano 3 Morellino Scansano

9 BM 1_5 CL 1 Col D'Orcia 4 Brunello di Montalcino

14 BM 6 Casanova Casanova di Neri 4 Brunello di Montalcino

15 BM 7 La Mannella La Mannella 4 Brunello di Montalcino

16 CC 1 Capannelle Capannelle 5 Chianti Classico

17 CC 2 Albola Albola 5 Chianti Classico

Table 9. Prospect of vineyards chosen for the statistical analysis of the soil.

Parameter M.U.

pH

Sand %

Silt %

Clay %

Total limestone %

Active limestone %

Cation exange capacity meq/100g

Electrical conducity dS/m

Total nitrogen g/Kg

Organic substance %

P205 ppm

CaO ppm

Mg0 ppm

K20 ppm

Table 10. Prospect of parameters chosen for the statistical analysis of the soil.

53

Code Sand Silt Clay pH

Electrical

conductivity

Cationic

exange

capacity

Organic

matter

CCP 1 28,0 44,0 28,0

8,50 0,11 14,50 0,97

MC 1 49,0 26,0 25,0 7,90 0,16 17,40 1,46

MC 2 49,0 26,0 25,0 7,90 0,16 17,40 1,46

MC 3 37,0 36,0 27,0 7,80 0,45 14,60 1,12

MC 4 58,0 22,5 19,5 7,90 0,15 13,40 1,17

MC 6 67,0 19,0 14,0 7,90 0,13 9,47 0,88

MC 7 70,0 20,0 10,0 6,80 0,07 15,40 1,20

SC 1 29,0 39,0 32,0 8,30 0,14 14,60 2,30

BM 1_5 14,0 60,0 26,0 8,30 0,13 15,66 0,27

BM 6 42,4 21,6 36,0 8,15 0,44 18,80 0,53

BM 7 57,0 20,0 23,0 8,41 0,30 18,40 1,29

CC 1 61,7 21,3 17,0 8,40 0,13 16,20 1,22

CC 2 49,6 23,0 27,4 8,40 0,13 25,30 1,36

Table 11. Physical characteristics of the experimental soils.

Code

Total

limestone

Active

limestone

Tot.

nitrogen P205 CaO Mg0 K20

CCP 1 21,00 14,50 0,06 10,80 3150 132 70

MC 1 5,90 1,90 0,09 4,00 3050 130 125

MC 2 5,90 1,90 0,09 4,00 3050 130 125

MC 3 39,70 8,00 0,07 5,00 2600 130 100

MC 4 0,22 2,85 0,05 4,50 2350 110 108

MC 6 23,50 3,80 0,06 5,00 1650 90 91

MC 7 0,70 0,10 0,10 21,50 1.684 490 181

SC 1 31,00 14,60 0,20 18,00 3650 320 141

BM 1_5 34,80 0,14 0,03 4,00 3400 185 90

BM 6 0,50 0,10 0,07 5,00 3212 1.135 324

BM 7 19,50 3,80 0,09 9,00 4796 181 98

CC 1 9,24 2,39 0,02 5,00 2760 161 410

CC 2 8,30 4,75 0,09 5,50 2930 154 322

Table 12. Chemical characteristics of the experimental soils.

54

Figure 39. Dendrogram obtained from the texture analysis of the soils.

The dendrogram obtained from the texture analysis (fig. 39) showed that the soils were

included into two large groups, in which we find in the first nine soils with a prevailing sandy

composition, while in the second one mostly silty soils (4 vineyards).

Figure 40. Dendrogram obtained from the physical and chemical analysis of the soils.

55

Examining chemical and physical characteristics of the soils was obtained the second

dendrogram (fig. 40).

The soil of the vineyard located in the „Chianti Colline Pisane‟ is a sandy, alkaline,

calcareous soil, poor in organic matter and potassium, with a mean amount of magnesium and

phosphorus (tab. 11-12).

The soils of the vineyards of „Montecucco‟ area have several differences between them, so as

they can be grouped into three clusters (tab. 11-12).

In the first one MC1 and MC 2, almost identical, are characterized by prevalence of sand,

sub-acid pH, moderate presence of limestone, low organic substance, and average potassium,

magnesium and exchange cation capacity. The MC3 and MC 4 even though appear in the

same cluster have some differences, for example the MC4 has greater content of sand while

the MC3 has equal proportions of sand, silt and clay. Different is the amount of total

limestone but not the active limestone which is low, in addition pH, organic matter and the

macro elements are at low or medium similar levels in both vineyards.

The third cluster is composed by MC5 and MC6 which have in common the high percentage

of sand and the low amount of clay. The MC5 soil has a slight range of active lime and a low

content of all the other elements. Soil of the vineyard MC6 has a lower pH, almost no

limestone, low organic substance and available nitrogen, meanwhile it has a good amount in

phosphorus, magnesium and potassium content.

The two vineyards located in the „Chianti Classico‟ belong to the same cluster although are

characterized by medium to high content of sand, medium silt and variable amount of clay.

Both soils are alkaline, moderately rich in active limestone, organic substance and mineral

elements (tab. 11-12).

The soil of the vineyard of „Magliano‟ has some similarities to one of the „Brunello di

Montalcino‟ (BM1_5) vineyard („Col D'Orcia‟ estate), for the prevalence of silty particles, the

same alkaline pH and quantity of active limestone, while the soil of „Magliano‟ estate is rich

in organic matter and well provided by all the other elements, while the BM1_5, is poor of

organic matter and by the other macro elements (tab. 11-12).

The soil of the other two vineyards of the „Montalcino‟ area are very different, in particular

BM6 („Casanova di Neri‟) is clay-sandy, poor in organic matter and phosphorus, while is rich

in potassium and showed an excess of magnesium. The soil of BM7 wich has a prevalence of

sand, is calcareous and rich in magnesium and poor in potassium.

56

3.5 The study of geopedologic, orographic and viticultural aspects:

the ColleMassari estate

As for data on soils (www.soilmaps.it) bibliographic data (Costantini et al.; 2006) and current

database was used as well as specific analyses conducted on the soils of the specific territory.

In depth study was carried out in the DOC „Montecucco‟ area where six vineyards from the

same wine farm were chosen, from which data on the geopedologic characteristics of the soils

were obtained from a previous study (Lizio 1999).

In this case the chemical-physical characteristics of the soil, were obtained from soil samples

analyzed in a special laboratory using methods approved by the „Società Italiana di Scienza

del Suolo‟. The geopedologic study refers to the land in the „ColleMassari‟ area, in the town

of „Cinigiano‟ (Gr) on a surface area of around 200 ha, where the six vineyards in question

are located. A general geological and geopedologic survey was carried out first. A study on

the soil type was carried out following the geological study of the area, using where possible a

manual drill to the depth of 80-100 cm, as often there were strata of compact pebbles and

clays. The limits between different soils are never clearly defined but the passage always

occurs through transition forms. The open profiles have been marked on the topographic map

to the scale of 1:5000, by enlarging the topographic base to the scale of 1:25.000 of the IGM.

During this geopedologic survey the World Reference Base for Soil Resources (WRB) was

used, which has led to the adoption of an innovative soil classification, already codified and

used internationally for geopedologic surveys. This type of classification regards the

functional characteristics of the soil as important without systematically subordinating it to

climatic data which obviously must be part of the interpretative aims of a project of zoning

but not necessarily be part of the geopedological definition of the interpretative model of the

soil examined and least of all subordinating it (Costantini and Lizio, 1996 ). The observations

transcriptions in the world soil classification (WRB) of the Fao, the last version of which was

published in 2001, conforms to the geoviticulture (Vaudour E., 2005) prospective. The

analysed data is always reinterpreted in the light of those modifying processes that occur on

the soil, therefore, it would be good practice to inspect the soils in different periods of the

year, as was done in this study. Studying the soil also means placing it in relation with the

landscape in which it is found so as to understand how the pedogenesis factors act, ie. the

climatic, biological, anthropical and geomorphologic processes occur on the territory

(Costantini and Lizio, 1996).

http://www.soilmaps.it/

57

For a better understanding of the pedologic data there is a list of suffixes used to describe the

soil strata that for the most part traces the indications given by the „Chiavi della Soil

Taxonomy ed.1992‟.

STRATA Ap: mineral strata that form on the surfaces interested by agricultural work

(trenching – ploughing).

STRATA B: strata that form below strata Ap and that are well structured .

HORIZON or STRATA C: strata that are not so influenced by pedogenetic processes and lack

the characteristics of strata Ap and B, but are made of hard rock.

HORIZON or STRATI R: presence of hard rock and impenetrable from the plant roots, the

rock is not possible to dig out with a spade.

Suffix of the horizons:

g (gley) hydromorphia linked to the drainage limitations in the soils, or to a saturation of the

horizons with stagnating water;

w (weathering) used to draw an alteration horizon B in which materials of soil origin are

differentiated by colour, structure or both;

t horizon of alluvial clay cumulus;

k horizon of calcium carbonate cumulus;

r (rock) symbol used to characterize the horizons C, made of soft rock, partially cemented; in

any case materials that can be dug up with a spade, but cannot be penetrated from the plant

roots save through their fractures;

n exchangeable sodium cumulus.;

AWC (usable water): differences between field capacity and withering point.

Water reserve classes ( A, W, C in mm):

Very high> 200 mm;

High 150-200 mm;

Moderate 100-150 mm;

Low 50-100 mm;

Very low < 50mm

In particular for this wine farm the study was carried out on soils from 4 specific vineyards

that are identified and described in the units they belong to. Their physical characteristics

have been determined by studying soil conduits, both by manual drill and by opening

pedologic profile, with samples per soil strata (Lizio, 1999).

58

3.5.1 ‘Campo La Mora’

The soil where the vineyard is located falls within the cartographic unit called „Unità Bocca

Nera‟ which occupies the moderate convex slopes , with excessive internal drainage.

Lithology: Polygenic conglomerates of the Messinian sup.. The substrata is made up of

pebbles and sand with areas where the pebble concentration prevails on the sandy part. The

soils present a sequence Ap/C, are moderately deep with a skeleton of 35% to 40%.

The horizon from 0 to 40 cm light yellowy brown in colour has a structure that tends to be

loose; an open sandy clayey texture, abundant skeleton, limestone, pH 7,9 ,a horizon from 40

to 110 cm pale brown in colour, bulky, open texture abundant skeleton; high in limestone and

pH 8,1. The organic substance is low and the C.E.C. average on all the pedologic profile (tab.

13).

The apparent density is 1,4 gr/cmc, useful water calculated AWC is 95mm, belonging to the

low water reserve (50-100mm). As for taxonomy the soils belong to the Xerorthents typical

open skeleton .

Figure 41. Ground‟s surface of „Campo la Mora’.

Figure 42. Soil‟s profile of „Campo la Mora’.

.

59

PHYSICAL AND

CHEMICAL

ANALYSIS VALUE JUDGMENT

Skeletron Ab. abundant

Sand 50%

Silt 32%

Clay 18%

pH 8,1 medium alkaline

Electrical conducity 0,118 mS normal

Total limestone 59,00% calcareous

Active limestone 7,4 medium

Organic matter 0,14 medium-low

NUTRIENTS


Total N 0,02% medium-low

Assimilable P 4 ppm medium-low

Assimilable Fe 3 ppm low

Assimilable Mn 8,2 ppm medium

Assimilable Cu 0,2 ppm medium-low

Assimilable Zn 0,3 ppm medium-low

Soluble Bo 0,26 ppm medium-low

Exchangeable Ca 1800 ppm high

Exchangeable Mg 140 ppm medium

Exchangeable K 80 ppm low

Exchangeable Na 100 ppm normal

C.E.C. ANALYSIS VALUE (meq/100 g) JUDGMENT

C.E.C. 10,80 meq medium

Ca 9 meq 83,3% high

Mg 1,17 meq 10,8% high

K 0,20 meq 1,9% low

Na 0,43 meq 4,0% normal

Basic saturation 100% high

mg/K ratio (meq/meq) 5,9 high

Table 13. Physical and chemical analysis of the soil of „Campo la Mora‟(profile 40-110 cm).

60

3.5.2 ‘Cerrete’

The soils relative to the vineyard of the same name are found between two cartographic units:

„Unità Poggi del Sasso‟ (fig. 46) and a second unit that represents a vertical discontinuity

between „Unità Colle Massari‟ and „Unità Poggio Formicone‟. The vineyard is situated on a

„rittochino‟ slope, the soil at the base of the plot is in the category of soils with discontinuous

characteristics inside the „Colle Massari‟ unit.

Such soils present probable water slump of contact between the pebbly part and the fine

reddish part. Contact situations exist between sediments containing pebbles up to 100/110 cm

and bulky reddish sediments.

Therefore there are notable differences between the deep strata and the superficial strata in

terms of texture, limestone content and physical - chemical analyses (tab. 14).

PHYSICAL AND

CHEMICAL ANALYSIS VALUE JUDGMENT

Skeletron Ma. marginal

Sand 42%

Silt 28%

Clay 30%

pH 0,3 alkaline



Active limestone 4,3 low


NUTRIENTS ANALYSIS VALUE JUDGMENT



Assimilable Fe 4,0 ppm low





Exchangeable Ca 2850 ppm medium-high

Exchangeable Mg 175 ppm high



61



Ca 14,25meq 87,30% high

Mg 1,46 meq 8,9% medium

K 0,20 meq 1,20% low

Na 0,43meq 2,60% normal


Table 14. Physical and chemical analysis of the soil of „Cerrete‟(profile 110-170 cm).

The soil upstream the plot falls within the „Poggi del Sasso‟ unit (fig. 45) and occupies the

sides with the low to moderate slopes badly drained. The substrata is made up of fine sandy

silts sediment with calcium carbonate concentrations right up to the surface.

Figure 43. Ground‟s surface of „Cerrete’.

62

Figure 44.Vertical discontinuity between

„Unità ColleMassari‟ and „Unità Poggio

Formicone‟.

The soils that fall in this category present a sequence Ap/CBgl/Cg2, are deep soils with a poor

skeleton and with obvious signs of hydromorphia. The horizon from 0 to 20/30 cm light olive

brown in colour, has an angular structure, moderately developed average, open texture,

ordinary skeleton, very chalky, pH 8, modest ordinary small red and grey mottles and small

ordinary limestone carbonate concretions. The horizon from 20/30 cm to 70 cm light olive

brown in colour, with average angular multifaceted structure, moderately developed, poor

skeleton, very calcareous, with pH 8,5, clear ordinary small red and grey mottles and small

ordinary limestone carbonate concretions. The horizon from 70 to 120 cm grey brown in

colour, lacking in structure and massive, open clayey texture, no skeleton, very calcareous,

with pH8, clear, ordinary small red and grey mottles, sodium content slightly high. The

organic matter is generally very low, average C.E.C., on all the pedologic profile. The

apparent density is 1,35 g/cmc for the superficial horizon, 1,3 g/cmc for the horizon below,

1,4 g/cmc for the third horizon; useful water calculated AWC is 168 mm, belonging to the

high water reserve class (150-200 mm).

Figure 45. „Poggi del Sasso‟ unit..

63

The taxonomy of the soils belong to the Aquic Xerorthens, open fine on fine clay with soda

clay.

Observations: as can be seen from the analysis the texture goes from sandy for the top soil to

clay silt for the sub soil, here too the depth increases electric conductibility associated to high

sodium content both in absolute value and in relation to C.E.C. .This is a soil characteristic of

the „Poggi del Sasso‟ that presents a finer texture than the other soils in the wine farm with a

presence of sodium in the sub soil associated to a higher clay content and to clear

hydromorphy, sign of imperfect drainage .

3.5.3 ‘Orto del prete’

The vineyard is located in a transition area between two cartographic units: „Unità Colle

Massari‟ and slight variation from the‟ Colle Massari‟ and „Bocca Nera‟ units.

„Colle Massari‟ occupies the steep slopes, with excessive internal drainage .

Lithology: Polygenic Messinian sup. conglomerates.

The substrata is made up of pebbles in sandy soil with areas in which the pebbly part prevails

on the fine sand, abundant presence of pebbles right from the surface. Such soils belong to the

category of soils that have a sequence Ac/C/Cr, moderately deep with a rich skeleton from

35% to 40%.

The horizon from 0 to 30/40 cm yellowy brown in colour , presents a structure that tends to

be loose, sandy texture, rich skeleton, very chalky with pH 7,9, the horizon 30/40 cm to 70/80

cm pale brown in colour, bulky, sandy franco texture, rich skeleton, with a pH 8,1; the strata

from 70/80 cm to 110 cm is very pebbly with thin layers of chalky sand.

The organic substance is very low and low C.E.C., on all the pedologic profile.

The nutrients analyses show a low quota of macro elements nitrogen, phosphorous and

potassium as well as inadequate secondary macro elements and microelements.

The apparent density is 1,4 gr/cmc, useful water calculated AWC is 72 mm, belonging to low

water reserve class (50-100 mm).

Regarding taxonomy the soils belong to Xerorthents typical sandy skeleton.

The „Bocca Nera‟ unit occupies the convex moderately steep slopes, with excessive internal

drainage.

Lithology: Messiniano sup. Polygenic conglomerates.

64

The substrata is made up of pebbles in sandy soils with areas in which the pebbly part prevails

on the fine sand. The soils have a sequence Ap/C are moderately deep soils with rich skeleton

from 35% to 40%.

The horizon from 0 to 40 cm light yellowy brown in colour has a structure that tends to be

loose, a clayey sandy texture, rich skeleton, chalky at pH 7,9, the horizon from 40cm to 110

cm pale brown in colour, bulky, with texture, rich skeleton, very chalky with pH 8,1 (tab. 15).

low organic matter and average C.E.C., on all the pedologic profile.

The apparent density is 1,4 gr/cmc, useful water calculated AWC is 95 mm, belonging to the

low water reserve class (50-100 mm).

As for taxonomy the soils belong to the Xerorthents typical skeleton.

Observation: for the soil that hosts the vineyard „Orto del Prete‟ the same is valid as for

„Campo la Mora‟ and for „Vigna Vecchia‟ in as much as the two cartographic units intersect

to which the soils of these two theses belong.

PHYSICAL AND

CHEMICAL



Sand 50%

Silt 32%

Clay 18%

pH 8,1 medium alkaline





NUTRIENTS




Assimilable Fe 3 ppm low






Exchangeable Mg 140 ppm medium



65



Ca 9 meq 83,3% high

Mg 1,17 meq 10,8% high

K 0,20 meq 1,9% low



Table 15. Physical and chemical analysis of the soil of „Orto del Prete‟(profile 0-120 cm).

Figures 46-47. Ground‟s surface and soil‟s profile of „Orto del Prete’.

3.5.4 ‘Vigna Vecchia’

The land on which the vineyard is located falls within the cartographic unit called „Unità

Colle Massari‟ which occupies the very steep slopes, with excessive internal drainage.

Lithology: Messinian sup. Polygenic conglomerates. The substrata is made up of pebbles in

sandy soils with areas in which the pebbly part prevails on the fine sand, abundant presence of

pebbles right from the surface. Such soils belong to the category that have a sequence

Ac/C/Cr, moderately deep with rich skeleton from 35% to 40%.

66

The horizon from 0 to30/40 cm yellowy brown in color , has a structure that tends to be loose,

a sandy texture, rich skeleton, very chalky at pH 7,9, the horizon from 30/40 cm to 70/80 cm

pale brown in colour, bulky, with sandy texture, rich skeleton, very chalky with pH 8,1, the

strata from 70/80 cm to 110 cm is characterized abundant pebbles with thin layers of chalky

sand.

The organic matter and C.E.C. are low, on all the pedologic profile.

The nutrients analyses moreover show a deficit in macro elements, nitrogen, phosphorous and

potassium as well as inadequate secondary macro elements and microelements (tab. 16).

The apparent density is 1,4 gr/cmc, useful water calculated AWC is 72 mm, belonging to the

low water reserve class (50-100 mm).

As for taxonomy the soils belong to the Xerorthents skeleton.

Figure 48. Ground‟s surface of „Vigna Vecchia‟.

Figure 49. Soil‟s profile of „Vigna Vecchia’.

67

PHYSICAL AND

CHEMICAL



Sand 67%

Silt 19%

Clay 14%

pH 7,9 sub alkaline





NUTRIENTS


Total N 0,06% low


Assimilable Fe 3,2 ppm low


Assimilable Cu 3,6 ppm medium


Soluble Bo 0,42 ppm low


Exchangeable Mg 90 ppm low




C.E.C. 9,47 meq low

Ca 5,25 meq 87,2% high

Mg 0,75 meq 7,9% medium

K 0,23 meq 2,4% medium



mg/K ratio (meq/meq) 3,3 medium

Table 16. Physical and chemical analysis of the soil of „Vigna Vecchia‟(profile 0-30/40cm).

68

Figure 50. Geopedologic card scale 1:5000; vineyard Campo la mora ( ), Vigna Vecchia ( ), Orto del Prete ( ), Cerrete (

Orto del

Prete

Vigna

Vecchia

Cerrete

Campo la Mora

69

3.6 Harvest date

Figure 51. Harvest date: comparison among 2009, 2010 and 2011 seasons.

Harvest time is set between the second half of September and the second half of October. For

the year 2009, the first area harvested was the „Chianti‟ from the „Colline Pisane‟, while the

last was the „Chianti Classico‟. Among the theses within the same production areas harvesting

time differs only a few days. The year 2010 recorded lower average temperatures compared

to the year before and this justifies the delay in harvesting in most of the thesis analysed,

exception made for the „Morellino di Scansano‟ Denomination and for the two belonging to

the „Chianti Classico‟, where it was necessary to delay the harvesting even more. Only the

bunches from the producer „Fattoria di Magliano‟ were picked in September and not in

October as in the case of the others. Early harvesting on the other hand, in all the farms in the

year 2011 was completed by the 20th

September. Such a result was no surprise because the

year 2011 was hotter in the period from April to end of October and therefore with a greater

cumulus of Growing Degree Days in all the areas. The bunches belonging to the „Chianti

Colline Pisane‟ denomination were harvested last while two years before they had been the

first to be harvested (fig. 51).

70

3.7 Sensorial characteristics of the grapes at harvest’s time

The MANOVA was conducted by examining the parameters linked to the sensorial analysis

of the grapes (tab. 17) at harvest‟s time harvested (using the values of three repeated analysis

of 17 theses each year for three years), studying the importance of the variables in function of

the chosen factor (tab. 18).

The real numeric values expressed by the panel were convert into percentages values relative

to the maximum value of 4.

Variable Abbreviation Units

Berry colour Ber. Col. s

Skin texture Skin text. s

Skin astringency Skin astr. s

Skin bitterness Skin bit. s

Skin aroma Skin aro. s

Skin maturity Skin mat. s

Pulp separation Pulp sep. s

Pulp acidity Pulp acid. s

Pulp swetness Pulp swe. s

Pulp aroma Pulp aro. s

Pulp maturity Pulp mat. s

Seed colour Seed col. s

Seed hardness Seed har. s

Seed bitterness Seed bit. s

Seed astringency Seed astr. s

Seed aroma Seed aro. s

Seed maturity Seed mat. s

Berry aroma Berry aro. s

Berry sensorial maturity* B.S.M. s

* Berry sensorial maturity was calculated by summing skin‟s, pulp‟s and seed‟s maturity

Table 17. List of abbreviations.

71

Factor Dependent

variable F Sig. Thesis Berry colour 25,727 ,000

Pulp separation 16,794 ,000

Pulp sweetness 8,436 ,000

Pulp acidity 4,875 ,000

Pulp aroma 7,100 ,000

Skin texture 11,149 ,000

Skin astringency 17,699 ,000

Skin aroma 7,943 ,000

Skin bitterness 10,120 ,000

Seed colour 9,094 ,000

Seed hardness 3,769 ,000

Seed bitterness 13,893 ,000

Seed astringency 6,431 ,000

Seed aroma 7,427 ,000

Pulp maturity 4,068 ,000

Skin maturity 9,189 ,000

Seed maturity 3,116 ,000

Berry aroma 2,277 ,007

Berry sensorial

maturity 1,554 ,096

Factor Dependent

variable F Sig. Year Berry colour 1,577 ,212













Seed aroma 126,072 ,000





Berry sensorial

maturity 12,986 ,000

Factor Dependent

variable F Sig. Thesis

* Year Berry colour 16,377 ,000


















Berry sensorial

maturity 2,315 ,001

Table 18 a, b, c. Test of the effects between subjects. (p=<0,05).

72

All the dependant variables proved to be statistically significant, bar the parameter indicated

with berry sensorial maturity linked to the sum of skin‟s, pulp‟s and seed‟s maturity. Such

variables do not change their level of significance if, the factor chosen during the statistical

analysis, is that of the thesis or the year or interaction between the two. However choosing the

year, a non significant statistical variable, the colour of the berry is added (tab. 18).

By using the data previously obtained the level of variability attributable to the different

factor was calculated (tab. 19). For most of the parameters the variability is attributable to the

year; the thesis, however, shows more variability as concerns berry colour and skin

astringency. Skin bitterness shows comparable levels of variability attributable to the different

source; most likely this is due to the difficulty in judging the sensation of bitterness.

Variable

%

variability

due to the

thesis

%

variability

due to the

year

%

variability

due to

interaction

thesis/year

%

variability

due to

error Berry colour 57,58 3,53 36,65 2,24 Pulp separation 13,13 73,38 12,71 0,78 Pulp sweetness 36,59 45,09 13,99 4,34 Pulp acidity 17,74 66,50 12,13 3,64 Pulp aroma 36,44 45,30 13,13 5,13 Skin texture 29,08 46,15 22,16 2,61 Skin astringency 45,03 20,63 31,79 2,54 Skin aroma 23,01 48,62 25,47 2,90 Skin bitterness 22,62 36,77 38,37 2,24 Seed colour 32,32 41,72 22,41 3,55 Seed hardness 5,11 79,33 14,20 1,36 Seed bitterness 11,42 77,91 9,86 0,82 Seed astringency 5,27 84,89 9,01 0,82 Seed aroma 5,15 87,49 6,66 0,69 Pulp maturity 18,83 66,39 10,15 4,63 Skin maturity 27,94 41,53 27,50 3,04 Seed maturity 4,29 86,02 8,32 1,38 Berry aroma 6,87 79,77 10,35 3,02 Berry sensorial maturity 8,70 72,73 12,97 5,60

Table 19. Level of variability attributable to the different factor.

73

From Tukey‟s statistic test it is possible to note the table with the different subsets and those

non differentiated; the most of significant variables creates differentiated homogenous

subsets, except for berry‟s aroma and sensorial maturity (tab. 20-22).

Regarding the sensorial maturity the values shown are the highest in one thesis of the „Chianti

Classico‟ area and the lowest in the one of „Morellino di Scansano‟ that is the thesis with less

difference as maturity level among the different parts of the berry (tab. 20). In general,

optimal value of sensory maturity are present in Siena‟s province; this area shows greater

variability in its theses if compared to Grosseto (tab. 20).

Grapes coming from „Chianti Colline Pisane‟ are characterized by differences among

maturation‟s level of the three parts of the berry; the skin appears more mature than seeds on

the contrary, seeds appear more mature than skin in one thesis belongs to the „Brunello di

Montalcino‟.

Analysing pulp‟s maturity, the lowest value belongs to the „Morellino di Scansano‟ thesis

while the highest to a „Brunello di Montalcino‟ thesis (tab. 20).

Regarding skin‟s maturity the highest value is found in one thesis of the „Brunello di

Montalcino‟ while the lowest value in „Morellino di Scansano‟ thesis already noted (tab. 21).

The values obtained from the analysis of seed‟s maturity indicate the lowest value in one

thesis of the „Chianti Colline Pisane‟ and the highest in one thesis of the „Chianti Classico‟

(tab. 22).

Code

Berry

colour Berry

aroma Berry

sensorial

maturity

Pulp

separation Pulp

sweetness Pulp

acidity Pulp aroma

Pulp maturity

CCP 1 93,06 bc 81,93 a 81,41 a 83,60 ab 88,9 b-e 84,73 a-d 83,89 a-e 85,15 ab

MC 1 98,61 ef 86,12 a 83,56 a 92,89 ef 93,12 de 84,73 a-d 92,28 ef 90,60 b

MC 2 99,33 f 81,93 a 83,62 a 90,35 c-f 94,79 e 86,40 cde 93,96 f 91,23 b

MC 3 99,33 f 79,13 a 83,28 a 92,04 def 91,4cde 87,24 cde 90,60 b-f 90,18ab

MC 4 96,23 c-f 79,41 a 84,24 a 86,98 a-d 92,2cde 83,89 a-d 88,08 b-f 87,66 ab

MC 5 98,75 ef 84,73 a 83,03 a 93,07 f 93,95 e 88,92 de 92,28 ef 92,07 b

MC 6 99,33 f 84,45 a 85,27 a 93,73 f 89,7cde 86,40 cde 87,24 b-f 89,13 ab

MS 1 86,05 a 78,57 a 85,47 a 90,35 c-f 78,02 a 78,85 ab 77,18 a 80,95 a

BM 1 90,16 b 83,89 a 86,25 a 81,91 a 89,76cde 83,89 a-d 87,24 b-f 85,57 ab

BM 2 96,34 c-f 85,01 a 86,34 a 87,82b-e 93,95 e 91,44 e 90,60 b-f 90,81b

BM 3 96,34 c-f 83,61 a 86,92 a 92,78ef 91,44cde 84,73 a-d 91,44 def 89,97ab

BM 4 95,60 cde 81,37a 87,25 a 86,13 abc 83,05abc 90,60 de 85,57 a-f 86,19 ab

BM 5 99,33 f 81,09 a 87,58 a 93,07 f 83,89 a-d 77,18 a 79,69 ab 83,47 ab

BM 6 98,31 def 85,29 a 88,32 a 93,73 f 93,12 de 88,08 de 93,12 ef 91,86b

BM 7 99,33 f 78,85 a 88,38 a 82,75 ab 88,08 b-e 82,21abc 87,24 b-f 84,93ab

CC 1 95,60 cde 78,29 a 88,81 a 87,82 b-e 79,69 ab 82,21abc 81,37 abc 82,63ab

CC 2 94,75 cd 84,91 a 89,01 a 84,44 ab 79,59 ab 78,76 ab 82,12 a-d 81,06 a

Table 20. Significant parameters with different and non differentiated subsets. Tukey (p=0,05).

74

Code Skin

texture Skin

astringency Skin

aroma Skin

bitterness Skin

maturity

CCP 1 91,44 g 91,44 g 72,14 a 88,92 fg 90,18 f

MC 1 88,08efg 83,89 c-g 76,34 ad 87,24 d-g 85,78 def

MC 2 81,37 a-e 75,50 bc 77,18 a-e 79,69 a-e 78,85 a-e

MC 3 72,98 a 65,43 a 73,82abc 72,98 a 71,09 a

MC 4 80,53 a-e 72,14 ab 77,18 a-e 76,34 abc 76,55abc

MC 5 90,60 fg 79,65 b-f 72,98 ab 78,85 a-d 83,05c-f

MC 6 87,24 dg 87,24 fg 77,18 a-e 88,92 fg 87,24 ef

MS 1 84,73 cg 77,18 bcd 78,85 a-e 83,89 c-f 80,74 b-e

BM 1 72,98 a 83,89 c-g 81,37 b-f 84,73 c-g 80,95 b-e

BM 2 78,85 ad 73,82 ab 81,37 b-f 83,05 b-f 78,23 a-d

BM 3 82,21 b-f 78,02 b-e 82,21 c-f 81,37 a-f 79,48 a-e

BM 4 77,18abc 86,40 efg 83,05def 88,08 efg 83,26 c-f

BM 5 90,60 fg 91,44 g 83,89def 93,12 g 90,18f

BM 6 84,73 cg 79,69 b-f 84,63def 83,89 c-f 82,42 b-f

BM 7 74,66 ab 71,30 ab 85,57 ef 74,66 ab 73,61 ab

CC 1 85,57 cg 75,50 bc 85,57 ef 76,34 abc 77,39 a-d

CC 2 84,62 cg 85,46 d-g 88,92 f 87,14 d-g 85,46 def


Code

Seed

colour Seed

hardness Seed

bitterness Seed

astringency Seed

aroma Seed

maturity

CCP 1 78,85abc 86,40 ab 67,11 a 74,66 ab 72,98 a 76,00 a

MC 1 79,69abc 88,08 ab 80,53 bcd 83,05 bc 82,21 be 82,71abc

MC 2 83,05bcd 84,73 ab 73,82abc 80,53 abc 72,98 a 79,02abc

MC 3 80,53abc 79,69 a 78,02 bcd 72,94 a 73,82ab 77,01ab

MC 4 85,57 be 79,69 a 72,98 abc 72,94 a 72,98 a 76,84 ab

MC 5 83,89bcd 89,76 b 78,86 bcd 79,69 abc 78,85 ad 82,21abc

MC 6 93,12 e 84,73 ab 77,18 bcd 77,18 ab 80,53 a-e 82,55abc

MS 1 85,57 be 86,40 ab 77,18 bcd 75,50 ab 81,37 a-e 81,20abc

BM 1 78,85abc 85,57 ab 80,53 bcd 78,02 ab 82,21 be 81,04abc

BM 2 85,57 be 86,40 ab 90,60 ef 87,24 c 87,24 de 87,41 c

BM 3 93,12 e 86,40 ab 81,37 cd 81,37 abc 83,0cde 85,06 bc

BM 4 89,76 de 88,92 ab 72,98 abc 75,50 ab 77,18abc 80,87abc

BM 5 87,24cde 87,24 ab 72,14 ab 79,69 abc 78,02abc 80,87abc

BM 6 84,73 be 90,60 b 80,53 bcd 78,02 ab 81,37 a-e 83,05abc

BM 7 78,02 ab 79,69 a 78,86 bcd 79,69 abc 75,50abc 78,35 ab

CC 1 79,69abc 79,69 a 83,05 de 82,21 bc 81,37 a-e 81,20abc

CC 2 72,90 a 83,79 ab 93,014 f 87,98 c 87,98 e 85,13 bc


75

From the multivariate analysis factor choice, the year and the statistic test it appears that most

of the parameters tested originate differentiated subsets, exception made for the variable

berry‟s colour. The variables linked to seeds are the variables that create well differentiated

subsets that indicate a great variability of data in the three years studied.

The year 2011 shows the highest parameters that influence the sensorial maturity of the

grapes at harvest‟s time especially for the variables correlated to the pulp, while the 2010

presents the lowest values that influence seed‟s level of maturity (tab. 23).

Variable 2009 2010 2011 Berry colour 95,98 a 96,70 a 96,11 a

Pulp separation 86,58 a 86,73 a 93,78 b

Pulp sweetness 86,60 a 87,49 a 91,47 b

Pulp acidity 82,60 a 83,05 a 88,51 b

Pulp aroma 84,68 a 87,93 b 89,25 b

Skin texture 79,35a 83,64 b 85,55 b

Skin astringency 77,72 a 80,07 ab 77,72 b

Skin aroma 76,68 a 81,72 b 81,99 b

Skin bitterness 79,35 a 84,53 b 84,81 b

Seed colour 80,53 a 84,96 b 85,12 b

Seed hardness 78,31 a 88,53 b 88,66 b

Seed bitterness 72,69 a 77,13 b 86,44 c

Seed astringency 74,02 a 75,94 a 87,62 b

Seed aroma 71,50 a 78,90 b 87,77 c

Pulp maturity 84,97 a 86,16 a 90,66 b

Skin maturity 78,27 a 78,27 b 78,27 b

Seed maturity 75,41 a 81,12 b 87,09 c

Berry aroma 78,71 a 81,77 b 86,34 c

Berry sensorial maturity 83,19 a 85,46 a 88,77 b

Table 23. Mean separation by multiple range test (Tukey); the comparison is among data shown in

horizontal.

In using the factorial statistic analysis – principal components method - it was possible to put

together all the variables noted and calculated in two new complex variables (components) so

as to represent 95,38% of the total variability of the sensorial characteristics of the berries at

harvest (tab. 24). The descriptors that represent the highest coefficients (tab. 25) operate in a

more reliable way in determining the characteristics of sensorial maturity of the berries at

harvest.

76

In particular the first component is mainly linked to the skin and pulp descriptors and the

second to the seed as seen in the value of the coefficients in order of increasing importance

(tab. 25).

Component

Weights of rotated factors

Total

%

variability

%

cumulated

1 10,72 59,54 59,54

2 6,45 35,84 95,38

Table 24. Results of the factorial analysis: principal components method.

Variable Component

1 2 Seed hardness ,999 ,048

Skin bitterness ,994

Seed maturity ,993

Seed astringency ,990

Seed aroma ,987

Skin aroma ,987

Skin maturity ,976

Seed colour ,971

Skin astringency ,962

Skin texture ,918

Berry aroma ,901 ,417

Seed bitterness ,560 ,489

Pulp acidity ,989

Pulp maturity ,996

Pulp aroma ,965

Pulp separ ,973

Berry colour ,973

Pulp sweetness ,995

Table 25. Matrix of the rotated components (Varimax) in order of importance. Coefficient values

below 0,4 are not included as of little relevance.

The average values of all the thesis per year obtained from the sensorial analysis and the

laboratory analysis (tab. 26) of the bunch macro structure at ripening were analyzed together,

to verify if there were any correlations among the different parameters noted.

77


pH pH n

Sugary content °Brix °Brix

Titratable acidity Titr. Ac. g/L

Ripening tecnological Index* Rip. T Index n

Bunch weight Bunch wgt g

Berry weight Berry wgt g

Berry seed number Berry seed num. n

Berry skin weight Skin wgt/berry g/berry

Berry seed weight Seed wgt/berry g/berry

Anthocyanins/berry Anth/berry mg/berry

Skins percentage % skin %

Skin anthocyanins Anth. skin mg/Kg

Skin polyphenols Polyph. skin mg/Kg

Skin polyphenols percentage % Skin polyph %

Seed polyphenols Polyph seed mg/Kg

Seeds percentage Seeds % %

Polyphenols/berry Polyph/berry mg/berry

Total Polyphenols Tot Polyph mg/Kg *Ripening Tecnological Index was so calculated °Brix/Titratable acidity


In using the factorial statistic analysis it was possible to put together all the variables noted

and calculated in three new complex variables so as to represent 99,84 % of the total

variability (tab. 27).

The first component is tied to the variables linked to skin‟s and seed‟s description; the third,

instead, to pulp‟s one.

The parameters that influence the technological ripeness and the phenolic richness the grapes

characterize the second component, one except for skin‟s anthocyanins, berry skin weight and

seed polyphenols tied to the third component (tab. 28).

The graph (fig. 52) shows how the descriptors that are in the same quadrant are directly

correlated while those further away are correlated negatively. The Ripening Index is indeed

positively correlated with berry‟s and skin‟s aroma, with seed bitterness and skin texture.

The first component is positively correlated with most of the variables examined, while the

second is correlated only in part.

The first and the second component are negatively correlated with the titratable acidity, the

skin‟s antochyanins, the mean number of seeds of the berry and the mean weight of the berry.

78

Component


Total

%

variability

%

cumulated

1 12,38 39,95 39,95

2 9,45 30,48 70,43

3 9,12 29,41 99,84


Component

1 2 3

Seed aroma ,977

Seed maturity ,969

Seed hardness ,940

Seed colour ,911

Berry aroma ,903


Skin aroma ,893 ,444

Skin bitterness ,893 ,441

Skin maturity ,863 ,497

Skin astringency ,840 ,525

Skin texture ,789 ,559

Seed bitterness ,781 -,402 ,477

Bunch weight ,619 ,500 -,605

Skin polyphenols ,577 -,800

pH ,411 ,911 ,042

Berry skin weight -,853

Ripening Index* ,923

Pulp aroma ,969

Pulp acidity ,977

Pulp maturity ,989

Pulp sweetness ,990

Berry colour ,963

Pulp separation ,963

Skin anthocyanins -,914

Berry weight -,918

Berry seed num -,925

Titratable acidity -,433 -,902

Total polyphenols -,517 -,835

°Brix -,580 -,726

Berry seed weight -,591 -,691 ,416

Seed polyphenols -,619 -,468 ,630



79

Figure 52. Rotated graph of the first two components obtained from the factorial analysis.

Analysing the multiple linear regression in the three new components and the sensorial

maturity of the berry, it is possible to note that there is a significant correlation and that the

total ripeness of the berry obtained experimentally is linearly correlated in a significant way

(r2 = 0,980) to that estimated in (tab. 29). Therefore, the berry sensorial maturity can be

expressed in the following way (tab. 30).

Berry sensorial maturity =86,337+F1*4,432+F2*3,923-F3*0,078

From the table of coefficient values for the estimation of the sensorial maturity of the berry, it

can be concluded that the model used is of significant importance and that the Berry sensorial

maturity variable, is more relevant with the first two constants rather than with the third one

(tab. 30).

80

Model R R-square R-square correct

Standard deviation Estimate’s error

1 ,990 ,980 ,980 ,970

Table 29. Statistic model of relation between the dependant variable of sensorial maturity of the berry

and the three new components obtained from factorial analysis.

Coefficients B

Standard

deviation

Error Sig. (Costant) 86,337 ,131 ,000

F 1 4,432 ,117 ,000

F 2 3,923 ,114 ,000

F 3 -0,078 ,074 ,295

Table 30. Coefficient values for the estimation of the sensorial maturity estimated of the berry,

standard error and their significance.

Calculating the relation between the berry sensorial maturity expressed and that estimated

statistically it can be seen graphically how the ripeness expressed by the panel and that

determined statistically are overlapping (fig. 53).

Figure 53. The relation between the ripeness index expressed and that statistically estimated.

81

In the table of correlations in the parameters examined in the course of our study, only the

variables characterized by a probability < 0,00001. Most of the variables show a positive

Pearson correlation value. The variable represented by the harvest date is linked negatively to

the most parameters analyzed except for titratable acidity.

As foreseen, the variables concerning the ripening of the single berry constituents, global and

sensorial ripening are strictly and positively linked to the single parameters that constitute

them. There aren‟t many significant correlation among parameters that influence the

technological ripeness and the phenolic richness and sensorial maturity (tab. 31).

82

Variable Ber.

col. Pulp

sep. Pulp

swe. Pulp

acid. Pulp

aro. Skin

tex. Skin

astr. Skin

aro. Skin

bit. Seed

col. Seed

hard. Seed

bit. Seed

astr. Seed

aro. Pulp

mat. Skin

mat. Seed

mat. Berry

aro. B. S.

M. Harv.

d. Berry colour 0,38 0,47 0,38 0,48 0,49 0,36 0,53

Pulp separation 0,38 0,60 0,48 0,55 0,51 0,37 0,46 0,40 0,42 0,75 0,39 0,47 0,58 0,64 -0,50

Pulp sweetness 0,47 0,60 0,69 0,89 0,42 0,46 0,45 0,37 0,93 0,45 0,67 0,72 -0,39

Pulp acidity 0,38 0,48 0,69 0,73 0,48 0,37 0,40 0,39 0,84 0,48 0,61 0,66 -0,42

Pulp aroma 0,48 0,55 0,89 0,73 0,44 0,38 0,39 0,33 0,92 0,45 0,69 0,70

Skin texture 0,27 0,51 0,62 0,76 0,64 0,41 0,55 0,40 0,40 0,82 0,46 0,63 0,70

Skin astringency

0,62 0,86 0,85 0,37 0,39 0,93 0,33 0,58 0,63

Skin aroma 0,37 0,42 0,32 0,76 0,86 0,81 0,47 0,54 0,37 0,44 0,94 0,46 0,76 0,76

Skin bitter 0,64 0,85 0,81 0,40 0,42 0,92 0,40 0,59 0,63

Seed colour 0,46 0,46 0,48 0,44 0,41 0,37 0,47 0,40 0,59 0,34 0,43 0,54 0,45 0,60 0,59 0,64

Seed hardness 0,32 0,45 0,38 0,55 0,39 0,54 0,42 0,59 0,44 0,51 0,64 0,44 0,52 0,77 0,71 0,67

Seed bitterness 0,37 0,44 0,83 0,89 0,39 0,85 0,66 0,59

Seed astringency

0,40 0,37 0,40 0,39 0,51 0,83 0,87 0,46 0,38 0,89 0,75 0,71 -0,46

Seed aroma 0,42 0,39 0,33 0,40 0,37 0,43 0,64 0,89 0,87 0,43 0,38 0,95 0,80 0,71 -0,44

Pulp maturity 0,49 0,75 0,93 0,84 0,92 0,40 0,44 0,54 0,44 0,39 0,46 0,43 0,38 0,55 0,75 0,81 -0,42

Skin maturity 0,39 0,34 0,24 0,82 0,93 0,94 0,92 0,45 0,52 0,15 0,38 0,38 0,38 0,45 0,70 0,75

Seed maturity 0,47 0,45 0,48 0,45 0,46 0,46 0,40 0,60 0,77 0,85 0,89 0,95 0,55 0,45 0,86 0,81 -0,43

Berry aroma 0,36 0,58 0,67 0,61 0,69 0,63 0,58 0,76 0,59 0,59 0,71 0,66 0,75 0,80 0,75 0,70 0,86 0,96 -0,41

Berry Sensorial

maturity 0,53 0,64 0,72 0,66 0,70 0,70 0,63 0,76 0,63 0,64 0,67 0,59 0,71 0,71 0,81 0,75 0,81 0,96 -0,37

Harvest date -0,50 -0,39 -0,42 -0,46 -0,44 -0,42 -0,23 -0,43 -0,41 -0,37

Table 31. Pearson‟s correlations. Legend abbreviations is on the previous pages.

83

The Stepwise discriminant analysis gave the best results compared to the traditional method;

analysis of the most relevant variables for statistics purposes were inserted in stepwise.

The 17 theses of our study, before being subjected to discriminant analysis were subdivided

by denomination area and company, thus creating a renumbering of the theses analyzed that

result be nine regarding this analysis (tab. 32).

Area Denomination area Company

1 Chianti Colline Pisane Beconcini

2 Montecucco Collemassari

3 Montecucco Salustri

4 Scansano Fattoria di Magliano

5 Montalcino Col D'Orcia

6 Montalcino Casanova di Neri

7 Montalcino La Mannella

8 Chianti Classico Capannelle

9 Chianti Classico Castello di Albola

Table 32. Area subdivision.

The discriminant analysis highlighted differences between areas examined and some of which may be

distinct (fig. 54). The first two canonical functions represent more than 76,0 % of the total


From the centroids graph obtained from the discriminant analysis (fig. 54), it is noted that the

points relative to the groups show a enough limited dispersion. Four distinct groups appear in

the centroid: the first and the second comprise theses coming from „Chianti Classico‟ area that

is well detached and the third, „Chianti Colline Pisane‟s grapes.

The fourth includes the residual theses: the numbers four and five are the more

distinguishable while many points linked to numbers two and seven are superimposed.

The two Montecucco theses do not appear very close differently from those of the „Brunello

and Montalcino‟. In the „Chianti Classico‟ the two theses are very different.

The 95,1% of the original grouping data were classified correctly, while 85,2% of the cases

grouped cross-validated are reclassified correctly (tab. 34).

84

Function Eingenvalue % of

variance %

cumulated Canonical

correlation 1 41,626 73,8 73,8 ,988

2 7,396 13,1 86,9 ,939

3 4,006 7,1 94,0 ,895

4 1,923 3,4 97,4 ,811

5 ,603 1,1 98,5 ,613

6 ,440 ,8 99,3 ,553

7 ,360 ,6 99,9 ,515

8 ,043 ,1 100,0 ,203

Table 33. Eigenvalues of discriminant analysis.

Figure 54. Centroids obtained from the cluster analysis of the sensorial characteristics of the grapes at

harvest‟s time.

Area

85

Area Group expected

Totals 1 2 3 4 5 6 7 8 9 C

ross

- V

alid

atio

n a

% 1 100,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0

2 ,0 100,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0

3 ,0 ,0 66,7 11,1 ,0 22,2 ,0 ,0 ,0 100,0

4 ,0 ,0 ,0 100,0 ,0 ,0 ,0 ,0 ,0 100,0

5 ,0 ,0 ,0 ,0 100,0 ,0 ,0 ,0 ,0 100,0

6 ,0 ,0 33,3 ,0 ,0 66,7 ,0 ,0 ,0 100,0

7 ,0 33,3 ,0 ,0 ,0 33,3 33,3 ,0 ,0 100,0

8 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0 ,0 100,0

9 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0 100,0



classified by the functions derived from all cases other than that case.

b.95,1% of original grouped cases correctly classified.

c.85,2% of cross-validated grouped cases correctly classified.

Statistical analysis was also used to investigate the features and possible statistically

significant differences of the grapes coming from the theses as part of the same Denomination

of Origin. The theses belonging to „Montecucco‟ area, were then subjected to multivariate

analysis, factorial, discriminating, linear regression and correlation. Among the theses coming

from the area of „Montalcino‟ was also studied the case of „Col D'Orcia‟ estate in order to

study in detail the clone effect grown in the same site of cultivation.

86

3.7.1 ‘Montecucco’ area

As results from the multivariate analysis of variance, statistically significant parameters do

not remain the same if the factor chosen during the statistical analysis is changed. The

parameters related to pulp, except for acidity and separation, remain no significant when the

source is represented by year or year interaction for thesis. If the year is chosen as the factor,

all the variables become significant (tab. 35).

Factor Dependent



Pulp sweetness ,824 ,541

Pulp acidity ,927 ,475











Pulp maturity ,649 ,664




Berry sensorial

maturity 1,834 ,131

Factor Dependent



















Berry sensorial

maturity 12,991 ,000

87

Table 35 a, b, c. Test of the effects between subjects (p=<0,05).

By using the data previously obtained, the level of variability attributable to the different


year; the thesis, however, shows more variability as concerns berry colour. Skin bitterness

shows comparable levels of variability attributable to the different factor.

It is noted a high percentage value of error in pulp aroma variable.

Factor Dependent




















Berry sensorial

maturity 1,396 ,222

88

Variable

%

variability

due to the

thesis

%

variability

due to the

year

%

variability

due to

interaction

thesis/year

%

variability

due to

error Berry colour 46,88 32,97 14,50 5,65 Pulp separation 6,08 91,21 1,69 1,02 Pulp sweetness 7,80 71,27 11,47 9,46 Pulp acidity 8,04 51,32 31,97 8,67 Pulp aroma 21,63 49,10 17,20 12,06 Skin texture 33,70 47,86 15,80 2,64 Skin astringency 35,09 43,65 19,71 1,56 Skin aroma 21,97 54,57 20,54 2,92 Skin bitterness 19,22 39,58 39,65 1,55 Seed colour 27,21 38,23 30,67 3,89 Seed hardness 9,47 71,83 16,87 1,83 Seed bitterness 5,88 77,92 14,31 1,89 Seed astringency 5,72 86,98 6,34 ,96 Seed aroma 5,75 86,97 6,34 ,94 Pulp maturity 6,32 78,46 5,48 9,74 Skin maturity 27,48 48,32 21,95 2,25 Seed maturity 4,89 84,72 8,48 1,91 Berry aroma 9,30 79,00 8,25 3,45 Berry sensorial maturity 10,65 75,44 8,11 5,81


From Tukey test it is possible to note the table with the different subsets and those non

differentiated (tab 37-39). The significant variable that creates the most differentiated

homogenous subsets is skin astringency. The values obtained from the analysis of the pulp,

instead, create no differentiated homogenous subsets.

„Montecucco‟ area shows optimal value of sensory maturity and small variability in its theses

(tab. 37-38). Regarding the sensorial maturity the values shown are higher in one thesis of

„ColleMassari‟ estate and in the thesis of „Salustri‟ estate (tab. 37).

The MC3 thesis is characterized by differences among maturation‟s level of the three berry

parts; the skin appears less matured than seeds and pulp; on the contrary, MC6 is

characterized by similar maturation‟s level of the berry at harvest‟s time (tab. 37-39).

Analysing pulp‟s maturity, the lowest value belongs to MC4 thesis while the highest to MC5

thesis (tab. 37).

Regarding skin‟s maturity the highest values are found in the grapes belonging to MC6, thesis

already highlighted as having similar maturation‟s level of the berry (tab. 38).

89

The values obtained from the analysis of seed‟s maturity indicate the lowest value in MC4

and the highest in MC1 (tab. 39).

The highest values are found in the pulp rather skin and seeds (tab. 37-39).

Code

Berry

colour Berry

aroma Berry

sensorial

maturity

Pulp

separation Pulp

sweetness Pulp

acidity Pulp aroma

Pulp maturity

MC 1 98,61 b 86,12 a 89,00 a 92,89 b 92,27 a 84,72 a 92,27 a 90,60 a

MC 2 99,33 b 81,93 a 86,34 a 90,35 ab 93,11 a 86,40 a 93,95 a 91,23 a

MC 3 99,33 b 79,13 a 83,61 a 92,04 b 91,43 a 87,24 a 90,60 a 90,18 a

MC 4 96,23 a 79,41 a 83,55 a 86,97 a 93,95 a 83,89 a 88,08 a 87,66 a

MC 5 98,75 b 84,72 a 88,37 a 93,06 b 89,76 a 88,92 a 92,27 a 92,06 a

MC 6 99,33 b 84,44 a 88,80 a 93,73 b 94,79 a 86,40 a 87,24 a 89,13 a


Code

Skin

texture Skin

astringency Skin

aroma Skin

bitterness Skin

maturity

MC 1 88,08 bc 83,89 de 83,89 bc 87,24 b 85,77 cd

MC 2 81,37 b 75,50 bc 78,85 abc 79,69 a 78,85 bc

MC 3 72,98 a 65,43 a 72,98 a 72,98 a 71,09 a

MC 4 80,53 ab 72,14 ab 77,17 ab 76,33 a 76,54 ab

MC 5 90,60 c 79,69 cd 83,05 bc 78,85 a 83,05 bcd

MC 6 87,24 bc 87,24 e 85,56 c 88,92 b 87,24 d


Code

Seed

colour Seed

hardness Seed

bitterness Seed

astringency Seed

aroma Seed

maturity

MC 1 79,69 a 88,08 b 80,53 b 83,05 b 82,21 c 82,71 a

MC 2 83,05 a 84,72 ab 73,82 ab 80,53 b 72,98 a 79,02 a

MC 3 80,53 a 79,69 a 78,01 ab 72,98 a 73,82 ab 77,01 a

MC 4 85,56 ab 79,69 a 72,98 a 72,98 a 72,98 a 76,84 a

MC 5 83,89 a 89,76 b 78,85 ab 79,69 ab 78,85 abc 82,21 a

MC 6 93,11 b 84,72 ab 77,17 ab 77,17 ab 80,53 bc 82,54 a


90

From the multivariance analysis it appears that all of the parameters tested originate

differentiated homogenous subsets. The variables linked to skin and seeds are the variables

that create well differentiated subsets that indicate a great variability of data in the three years

studied.

The year 2011 shows the highest parameters that influence the sensorial maturity of the

grapes at harvest‟s time especially for the variables correlated to the pulp; the 2009, instead,

presents the highest values that influence skin‟s level of maturity.

In all three years variables related to seed‟s maturity were those with the lowest values (tab.

40).

Variable 2009 2010 2011 Berry colour 99,33 b 97,90 a 98,55 ab

Pulp separation 95,00 b 83,60 a 95,93 b

Pulp sweetness 94,79 b 88,08 a 94,79 b

Pulp acidity 85,56 ab 83,46 a 89,76 b

Pulp aroma 89,34 ab 88,92 a 93,95 b

Skin texture 86,40 b 77,17 a 86,82 b

Skin astringency 83,47 c 70,88 a 77,59 b

Skin aroma 84,30 b 74,24 a 82,21 b

Skin bitterness 84,72 b 73,40 a 83,89 b

Seed colour 87,24 b 79,69 a 85,98 b

Seed hardness 88,92 b 75,08 a 89,34 b

Seed bitterness 76,33 b 69,62 a 84,72 c

Seed astringency 76,33 b 67,11 a 89,76 c

Seed aroma 78,01 b 65,01 a 87,66 c

Pulp maturity 91,01 b 85,88 a 93,53 b

Skin maturity 84,72 b 73,92 a 82,63 b

Seed maturity 81,37 b 71,30 a 87,49 c

Berry aroma 83,88 b 76,06 a 87,94 b

Berry sensorial maturity 88,33 b 81,23 a 90,28 b


horizontal.


and calculated in three new complex variables (components) so as to represent 98,95 % of the

total variability of the sensorial characteristics of the berries at harvest (tab. 41). The

descriptors that represent the highest coefficients (tab. 42) operate in a more reliable way in

determining the characteristics of sensorial maturity of the berries at harvest.

91

The first component is tied to the most of the parameters correlated to the pulp except for the

value of separation, to the berry‟s characteristics, to berry sensorial maturity and to the most

of the parameters correlated to the skin. Variables linked to the description of the seeds are

associated with the second component. The last component is characterized, however, by

skin‟s texture and bitterness (tab. 42).

Component


Total

%

variability

%

cumulated 1 8,21 43,22 43,22

2 6,82 35,88 79,10

3 3,77 19,85 98,95


Variable Component

1 2 3 Pulp aroma ,998

Skin aroma ,970

Pulp sweetness ,943

Pulp acidity ,924


Skin Astringency ,842 -,528


Berry colour ,767

Berry Sensorial maturity ,750 ,498 ,434

Seed colour ,697 -,421 ,579


Seed hardness ,927

Seed maturity ,890


Skin texture ,503 ,846


Seed aroma ,870 ,493





92

Extracting only two components from all the data collected it is possible to explain 85,02% of

the total variability (tab. 43). In particular the first component is greatly linked to the variables

already mentioned for the description of the first component; the second is composed of the

sum of the second and the third component of the previous analysis (tab. 44).

Component


Total

%

variance

%

cumulated 1 8,41 44,26 44,26

2 7,74 40,76 85,02

Table 43. Results of the factorial analysis extracting only the first 2 components: method of the

principal components.

Variable Component

1 2

Skin aroma ,986

Pulp aroma ,973

Pulp acidity ,897

Pulp sweetness ,875

Skin astringency ,849 -,513


Seed colour ,840

Skin maturity ,819


Berry colour ,792 ,435

Berry Sesorial maturity ,772 ,631

Skin texture ,757

Seed hardness ,833

Seed maturity ,979



Seed aroma ,984


Seed bitterness ,978

Table 44. Matrix rotated (Varimax) of the first two components extracted. Descriptors ordered in

order of importance excluding the <0,4 coefficients as of little relevance.

93


laboratory analysis of the bunch macro structure at ripening were analyzed together, to verify

if there were any correlations among the different parameters noted.




The first component is tied to some sensorial variables related to seeds and skin and some

technological and phenolic richness like sugary and anthocyanins content, the pH and the

mean weight of the berry. Variables linked to berry sensorial maturity, to skin astringency and

to the description of the pulp are associated with the second component.

The parameters that influence the seeds and titratable acidity characterize the third component

(tab. 46).

Component


Total

%

variability

%

cumulated

1 12,07 37,73 37,73

2 10,01 31,29 69,02

3 9,70 30,32 99,34


94

Variable Component

1 2 3 °Brix ,991

Berry seed number ,989

Seed polyphenols ,988

Ripening Tecnological Index ,987

pH ,885 -,438

Total polyphenols ,867 ,495

Berry weight ,625 -,763

Skin astringency ,566 ,622 -,540

Titratable acidity ,564 -,809

Berry skin weight ,484 ,874

Pulp aroma ,903

Skin aroma ,876

Pulp sweetness ,920

Skin polyphenols ,966

Pulp maturity ,913

Pulp acidity ,976

Berry seed weight ,991

Seed hardness ,469 ,883


Berry aroma ,975

Berry colour ,893

Seed colour ,738 -,663

Berry Sensorial maturity ,941

Skin maturity ,851

Seed astringency -,422 ,856

Seed maturity -,531 ,520 ,669

Seed aroma -,678 ,639

Seed bitterness -,706 ,667

Skin texture -,817 ,553


Skin bitterness -,909

Bunch weight -,992



95


the total variability (tab 47).

In particular the first component is greatly linked to the variables already mentioned for the

description of the first component; the second is composed of the sum of the second and the

third component of the previous analysis (tab. 48).

The graph (fig. 55) shows how the descriptors that are in the same quadrant and that are close

to the ripening of the berry are directly correlated to it while those further away are correlated

negatively. The Ripening Technological Index is indeed positively correlated with the sugary

content, the pH, the seed‟s and skin‟s weight, with the grade of polyphenols in the berry and

in the skins and, regarding sensorial parameters, with some characteristics of the pulp.

Berry sensorial maturity, on the contrary, is positively correlated, obviously with the level of

maturity of the three different part of the berry, berry‟s aroma and colour. It is negatively

correlated with the variables linked to technological and phenolic richness of the grapes at

harvest‟s time.

The first and the second component are negatively correlated with the mean weight of the

bunch.

Component


Total

%

variance

%

cumulated 1 13,25 41,39 41,39

2 10,52 32,87 74,26



96

Variable Component

1 2 pH 1,000


Berry weight ,912

°Brix ,912

Titratable acidity ,878


Ripening Tecnological

Index ,842


Skin aroma ,596 ,668

Pulp aroma ,567 ,777

Total Polyphenols ,561

Pulp sweetness ,942

Seed colour

Pulp acidity ,965

Pulp maturity ,983

Berry skin weight ,350

Berry colour ,846

Berry aroma ,963

Skin maturity ,602

Berry Sensorial maturity ,946


Seed hardness ,795


Berry seed weight

Skin bitter -,654

Seed maturity -,675 ,732

Seed astringency -,697 ,613

Skin texture -,701 ,538

Seed aroma -,818 ,572

Seed bitterness -,876 ,468

Bunch weight -,903

Skin anthocyans -,998



97


Analysing the multiple linear regression in the three new components and the berry sensorial

maturity, it is possible to note that there is a significant correlation and that the total ripeness

of the berry obtained experimentally is linearly correlated in a significant way (r2 = 0,993) to

that estimated (tab. 49). Therefore, the berry sensorial maturity can be expressed in the

following way (tab. 50).

Berry sensorial maturity =88,884+F1*3,571+F2*2,335+F3*2,066


can be concluded that the model used is of significant importance, that the Berry sensorial

maturity variable, is relevant with all three constants (tab. 50).

98



1 ,996 ,993 ,993 ,614



Coefficients B

Standard

deviation


F 1 3,571 ,115 ,000

F 2 2,335 ,122 ,000

F 3 2,066 ,086 ,000



Calculating the relation between berry sensorial maturity expressed and that estimated

statistically it can be seen graphically how the ripeness expressed by the panel and that



99

Examining the correlation between the parameters analyzed, in the table only the variables

with a probability <0,00001 are reported. The mean weight of the bunch and the skin‟s

anthocyanins, are the parameters that don‟t present significant Pearson‟s correlations (tab. 51).

In most cases there are values of Pearson's correlation positive, however the parameters

connected with the technological maturity of the grapes are negatively correlated. The

variable represented by the harvest date is linked negatively to the most parameters analyzed.



them. Positive Pearson correlation values were found in mean weight of the skins and

separation of the pulp. There aren‟t many significant correlation among parameters that

influence the technological ripeness and the phenolic richness and sensorial maturity.

Nevertheless positive Pearson correlation values were found in mean weight of the skins and

separation of the pulp and in skin‟s polyphenols and seed‟s astringency (tab. 51).

100

Ber.

col.

Pulp

sep.

Pulp

swe.

Pulp

acid.

Pulp

aro.

Skin

tex.

Skin

astr.

Skin

aro.

Skin

bit.

Seed

col.

Seed

hard.

Seed

bit.

Seed

astr.

Seed

aro.

Pulp

mat.

Skin

mat.

Seed

mat.

Berry

aro.

B. S.

M. Harv.d

Berry colour 0,60

Pulp

separation

0,65 0,51 0,58 0,65 0,69 0,68 0,78 0,75 0,76 0,75 0,75 -0,66

Pulp

sweetness

0,65 0,79 0,73 0,64 0,63 0,89 0,70 0,75 0,78

Pulp acidity 0,71 0,83 0,58

Pulp aroma 0,79 0,71 0,89 0,67 0,64

Skin texture 0,58 0,83 0,86 0,81 0,66 0,64 0,66 0,76 0,92 0,74 0,84 0,85

Skin

Astringency

0,83 0,91 0,80 0,61 0,67 0,44 0,94 0,67 0,79 0,80

Skin aroma 0,86 0,91 0,84 0,60 0,70 0,58 0,96 0,69 0,88 0,87

Skin

bitterness

0,81 0,80 0,84 0,59 0,62 0,59 0,72 0,93 0,71 0,74 0,77

Seed colour 0,60 0,58 0,62

Seed hardness 0,65 0,73 0,66 0,61 0,59 0,72 0,82 0,83 0,58 0,65 0,90 0,76 0,79

Seed

bitterness

0,69 0,64 0,62 0,72 0,87 0,88 0,63 0,61 0,88 0,77 0,79 -0,65

Seed

astringency

0,68 0,64 0,66 0,59 0,82 0,87 0,92 0,66 0,63 0,94 0,84 0,82 -0,79

Seed aroma 0,78 0,63 0,76 0,67 0,70 0,72 0,83 0,88 0,92 0,64 0,76 0,97 0,91 0,88 -0,76

Pulp maturity 0,75 0,89 0,83 0,89 0,58 0,58 0,63 0,66 0,64 0,69 0,80 0,82

Skin maturity 0,92 0,94 0,96 0,93 0,65 0,61 0,63 0,76 0,48 0,75 0,86 0,87

Seed maturity 0,76 0,74 0,67 0,69 0,71 0,58 0,90 0,88 0,94 0,97 0,69 0,75 0,90 0,91 -0,69

Berry aroma 0,75 0,75 0,67 0,84 0,79 0,88 0,74 0,76 0,77 0,84 0,91 0,80 0,86 0,90 0,98 -0,61

Berry

S.maturity

0,60 0,75 0,78 0,58 0,64 0,85 0,80 0,87 0,77 0,62 0,79 0,79 0,82 0,88 0,82 0,87 0,91 0,98

Harvest date -0,66 -0,65 -0,79 -0,76 -0,69 -0,61


101


analysis of the most relevant variables for statistics purposes were inserted in stepwise.

The discriminant analysis highlighted differences between areas examined and some of which may be

distinct (fig. 57). The first two canonical functions represent more than 91,6 % of the total



variance %

cumulated Canonical

correlation 1 42,273 82,6 82,6 ,988

2 4,788 9,4 91,9 ,910

3 2,602 5,1 97,0 ,850

4 1,103 2,2 99,2 ,724

5 ,430 ,8 100,0 ,549



points relative to the groups show a enough limited dispersion. Four distinct groups appear in

the centroid: the fifth includes the thesis coming from a different estate and one of „ColleMassari‟

winery indeed some points are superimposed.

The 95,1% of the original data grouping were classified correctly, while 92,6% of the cases


102


harvest‟s time.

Thesis

Group expected

Totals 1 2 3 4 5 6

Cro

ss-v

alid

atio

n a

% 1 88,9 ,0 ,0 ,0 ,0 11,1 100,0

2 ,0 100,0 ,0 ,0 ,0 ,0 100,0

3 ,0 ,0 100,0 ,0 ,0 ,0 100,0

4 ,0 ,0 ,0 100,0 ,0 ,0 100,0

5 ,0 ,0 ,0 ,0 66,7 33,3 100,0

6 ,0 ,0 ,0 ,0 ,0 100,0 100,0






103

3.7.2 ‘Col d’Orcia’ estate

From the multivariate analysis, with the thesis as factor, what emerges is that among the

dependent variables the no significant ones for the sensorial variables are seed‟s hardness,

berry‟s aroma and berry‟s sensorial aroma.

The variables analyzed change their level of significance if, the factor chosen during the

statistical analysis changes; indeed choosing the year, many non significant statistical variable

are added. In this case, separation, sweetness, aroma and maturity of the pulp and skin texture

become the only significant variables.

Pulp maturity and berry sensorial aroma are the two parameters that become non significant

when the factor is represented by year interaction by thesis (tab. 54).

Factor Dependent

















Seed maturity ,257 ,775


Berry sensorial

maturity 1,478 ,244

Factor Dependent



















Berry sensorial

maturity ,569 ,687

104


The variability quota attributed to the different factor was calculated using the data previously

obtained (tab. 55).

For most of the parameters the variability is attributable to the year; nevertheless most

variables show comparable levels of variability attributable to the different factor.

It is noted a high percentage value of error in berry sensorial aroma variable.

Factor Dependent




















Berry sensorial

maturity 1,946 ,089

105

Variable

%

variability

due to the

thesis

%

variability

due to the

year

%

variability

due to

interaction

thesis/year

%

variability

due to

error Berry colour 21,05 60,68 16,81 1,46 Pulp separation 14,98 81,40 2,96 ,66 Pulp sweetness 33,77 44,12 16,68 5,42 Pulp acidity 55,77 13,39 25,13 5,71 Pulp aroma 28,27 54,37 12,83 4,53 Skin texture 51,04 17,89 27,53 3,54 Skin astringency 57,97 4,75 33,40 3,88 Skin aroma 38,61 21,47 31,83 8,10 Skin bitterness 33,96 10,59 50,01 5,44 Seed colour 51,56 16,38 25,65 6,41 Seed hardness 4,00 25,39 61,90 8,71 Seed bitterness 50,27 1,14 45,91 2,68 Seed astringency 26,18 9,83 59,91 4,08 Seed aroma 27,26 6,21 61,43 5,10 Pulp maturity 18,09 62,99 12,22 6,70 Skin maturity 51,50 5,11 36,06 7,32 Seed maturity 23,21 2,10 66,53 8,17 Berry aroma 13,32 34,04 37,76 14,88 Berry sensorial maturity 11,40 29,60 38,97 20,03


From Tukey‟s test it is possible to note the table with the different subsets and those non

differentiated (tab. 56-58).

The significant variables create differentiated homogenous subsets except for seed‟s hardness

and its maturity, pulp‟s maturity, berry aroma and berry sensorial maturity.

Skin astringency is the variable that creates well differentiated subsets that give rise to a wide

range of data variability (tab. 56).

„Col d‟Orcia‟ grapes in general shows optimal values of sensory maturity; the highest values

are found in the pulp rather than in seeds and in skin; the latter is characterized by greater

variability (tab. 56-58).

BM2 thesis is characterized by the highest maturity of the pulp and seeds, on the contrary in

BM5 is skin to be more mature than the other two parts of the berry and it shows the lowest

values of pulp‟s and seed‟s maturity.

106

Code

Berry

colour Berry

aroma Berry

sensorial

maturity

Pulp

separation Pulp

sweetness Pulp

acidity Pulp aroma

Pulp maturity

BM 1 90,16 a 83,89 a 84,24 a 81,91 a 89,76 abc 83,89 ab 87,24 ab 85,56 a

BM 2 96,34 bc 85,00 a 87,58 a 87,82 b 93,95 c 91,43 b 90,60 b 90,81 a

BM 3 96,34 bc 83,61 a 86,92 a 92,77 c 91,43 bc 84,72 ab 91,43 b 89,97 a

BM 4 95,60 b 81,37 a 85,46 a 86,13 b 83,05 a 90,60 b 85,56 ab 86,19 a

BM 5 99,33 c 81,09 a 87,25 a 93,06 c 83,89 ab 77,17 a 79,69 a 83,47 a


Code

Skin

texture Skin

astringency Skin

aroma Skin

bitterness Skin

maturity

BM 1 72,98 a 83,89 bc 82,21 ab 84,72 a 80,95 a

BM 2 78,85 ab 73,82 a 77,17 a 83,05 a 78,22 a

BM 3 82,21 b 78,01 ab 76,33 a 81,37 a 79,48 a

BM 4 77,17 ab 86,40 cd 81,37 ab 88,08 ab 83,26 ab

BM 5 90,60 c 91,43 d 85,56 b 93,11 b 90,18 b

Table 57. Significant parameters with different subsets. Tukey (p=0,05).

Code

Seed

colour Seed

hardness Seed

bitterness Seed

astringency Seed

aroma Seed

maturity

BM 1 78,85 a 85,56 a 80,53 b 78,01 a 82,21 ab 81,03 a

BM 2 85,56 ab 86,40 a 90,60 c 87,24 b 87,24 b 87,41 a

BM 3 93,11 b 86,40 a 81,37 b 81,37 ab 83,05 ab 85,06 a

BM 4 89,76 b 88,92 a 72,98 a 75,50 a 77,17 a 80,86 a

BM 5 87,24 b 87,24 a 72,14 a 79,69 a 78,01 a 80,86 a


From the MANOVA analysis, it appears that most of the parameters examined originate non

differentiated subsets.

The variables linked to pulp are the variables that create well differentiated subsets that

indicate a great variability of data in the three years studied (tab. 59).

In 2010 the highest values were recorded concerning sensorial analysis, that show, at harvest,

a high maturity level of the berry in most of its components, compared to the other years

except for seed‟s variables and separation and acidity of the pulp (tab. 59).

107

The year 2011 stands out for the highest parameters that influence the description of seeds and

the 2009 just for seed hardness (tab. 59).

Variable 2009 2010 2011 Berry colour 96,98 b 99,05 b 90,62 a

Pulp separation 78,53 a 93,22 b 93,26 b

Pulp sweetness 83,55 a 91,10 b 90,60 b

Pulp acidity 84,56 a 84,05 a 88,08 a

Pulp aroma 82,04 a 92,11 b 86,57 a

Skin texture 77,51 a 83,55 b 80,03 ab

Skin astringency 83,05 a 84,05 a 81,03 a

Skin aroma 79,02 a 83,05 a 79,52 a

Skin bitterness 86,07 a 84,05 a 88,08 a

Seed colour 85,56 a 89,59 a 85,56 a

Seed hardness 88,58 a 84,05 a 88,08 a

Seed bitterness 79,02 a 79,02 a 80,53 a

Seed astringency 78,01 a 81,03 a 82,04 a

Seed aroma 81,54 a 80,03 a 83,05 a

Pulp maturity 82,04 a 89,97 b 89,59 b

Skin maturity 81,41 a 83,67 a 82,16 a

Seed maturity 82,54 a 82,74 a 83,85 a

Berry aroma 80,86 a 85,06 a 83,05 a

Berry sensorial maturity 84,80 a 88,22 a 85,85 a


horizontal.

By factorial analysis it was possible to reduce the number of variables to three main

components able to represent 98,88% of the total variability of the sensorial characteristics of

the berries at harvest (tab. 60).

In particular the first component is mainly linked to the skin and pulp descriptors and the

second to the seeds. Pulp acidity and seed astringency are associated with the third component

(tab. 61).

108

Component


Total

%

variability

%

cumulated 1 9,79 51,51 51,51

2 6,42 33,77 85,28

3 2,58 13,60 98,88


Variable

Component

1 2 3

Seed hardness ,989

Seed colour ,989

Seed maturity ,949




Seed astringency ,675 ,701

Skin aroma ,997

Skin maturity ,986

Pulp maturity ,970

Skin texture ,961

Pulp sweetness ,948


Pulp aroma ,919



Berry Sensorial maturity ,821 ,494

Berry colour ,687 ,427 ,410

Pulp separation ,643 -,704




the total variability (tab. 62). In particular the first component is greatly linked to the variables

that describe pulp‟s and skin‟s level of maturation; the second is connected to the seed‟s

characteristics and to the pulp acidity (tab. 63).

Component


Total

%

variance

%

cumulated 1 9,72 51,18 51,18

2 7,35 38,66 89,84



109

Variable Component

1 2 Seed maturity ,993

Seed aroma ,978


Seed hardness ,934

Seed colour ,934


Pulp acidity ,762

Skin aroma ,998

Skin maturity ,992

Pulp maturity ,981



Skin texture ,937

Pulp sweetness ,918

Pulp aroma ,884

Berry Sensorial maturity ,811 ,583


Pulp separation ,697 -,543

Berry colour ,689 ,563




laboratory analysis of the bunch (macro structure) at ripening were analyzed together, to

verify if there were any correlations among the different parameters noted.




The first component is tied to some sensorial variables related to seeds and pulp and to some

technological and phenolic richness like polyphenols total content, mean weight and mean

number of seeds. Variables linked to only sensorial variables are associated with the second

component.

The parameters that influence technological and phenolic characteristics, pulp acidity, seed‟s

aroma and astringency characterize the third component (tab. 65).

110

Component


Total

%

variability

%

cumulated

1 11,60 35,15 35,15

2 11,51 34,88 70,03

3 9,67 29,29 99,32


Variable Component

1 2 3

Skin polyphenols -,988



°Brix -,970

Berry weight ,912

Pulp separation -,887

Total polyphenols -,848 -,502


Seed colour ,836 ,494


Seed maturity ,702 ,532 ,472

Berry sensorial maturity ,989

Berry aroma ,953


Pulp maturity ,938

Skin maturity ,931

Skin texture ,908 -,416

Skin bitter -,458 ,883

Berry colour ,882

Skin aroma ,865

Pulp sweetness ,820 -,572

Seed bitterness ,526 ,742 ,414

Pulp aroma ,736 -,676

pH ,936

Bunch weight ,923

Ripening tecnological Index -,903

Skin anthocyanins -,405 -,887


Titratable acidity ,482 ,848

Seed polyphenols -,529 -,821

Harvest date ,695 -,712





111


the total variability (tab. 66).

In particular the first component is greatly linked to the variables that describe technological,

phenolic, sensorial characteristics especially regarding seeds except the sensation of

bitterness. The second is composed of only sensorial variables (tab. 67).



negatively. Berry sensorial maturity is positively correlated with aroma and colour of the

berry, maturity and sweetness of the pulp and with texture of the skin. It is negatively

correlated, on the contrary, with the variables linked to technological and phenolic richness of

the grapes at harvest‟s time especially regarding pH and mean weight of the berry and of the

bunch.

Ripening Technological Index is positively correlated with variables linked to technological

and phenolic characteristics and to separation of the pulp.

The first and the second component aren‟t negatively correlated with any variables.

Component


Total

%

variance

%

cumulated 1 16,19 49,06 49,06

2 11,76 35,65 84,70



112

Variable Component

1 2 Berry sensorial maturity ,979

Berry aroma ,977

Skin texture ,956

Skin maturity ,933

Pulp maturity ,928


Pulp sweetness ,889


Berry colour ,850

Pulp aroma ,820

Harvest date


Seed aroma ,952

Berry weight ,947




Seed maturity ,858 ,513

Bunch weight ,815


Seed colour ,814 ,517

pH ,790



Pulp acidity ,616

Skin aroma -,461 ,887

Pulp separation -,813

Skin polyphenols -,842

Ripening tecnological index -,844

Skin Anthocyanins -,864

°Brix -,887

Seed polyphenol -,919

Total polyphenols -,971



113


Analyzing the multiple linear regression between the three new components and the berry

sensorial maturity it was clear that an important correlation existed and that the berry

sensorial maturity experimentally determined is linearly correlated in an important way (r2 =

0,967) to that estimated (tab. 68). Thus the total ripening index of the berry can be expressed

as follows (tab. 69).

Berry sensorial maturity = 87,575+F1*1,182+F2*3,910-F3*0,012


can be concluded that the model used is of significant importance, however it is more

relevant with the component 1 and 2 rather than with 3 one (tab. 69).

114



1 ,984 ,969 ,967 1,092



Coefficients B

Standard

deviation


F 1 1,182 ,257 ,000

F 2 3,910 ,116 ,000

F 3 -,012 ,223 ,956



Calculating the relation between berry sensorial maturity expressed and that estimated

statistically it can be noted graphically how the ripeness expressed by the panel and that



115

In the table of correlations in the parameters examined in the course of our study, only the

variables characterized by a probability < 0,00001 are found on the table (tab. 70). The pH

and harvest date are the parameters that don‟t present significant Pearson‟s correlations .Most

of the variables show a positive Pearson‟s correlation value. Negative Pearson‟s correlation

values were found in many parameters linked to the weight and among technological and

phenolic characteristics.



them. There aren‟t many significant correlation among parameters that influence the

technological ripeness, and the phenolic richness and sensorial maturity. Nevertheless,

positive Pearson‟s correlation values were found between °Brix and skin‟s texture and its

maturity and negative between berry colour and berry seeds weight and pulp separation.

116

Variable Ber.

col.

Pulp

sep.

Pulp

swe.

Pulp

acid.

Pulp

aro.

Skin

tex.

Skin

astr.

Skin

aro.

Skin

bit.

Seed

col.

Seed

hard.

Seed

bit.

Seed

astr.

Seed

aro.

Pulp

mat.

Skin

mat.

Seed

mat.

Berry

aro.

B. S.

M.

Berry colour 0,65

Pulp

separation

0,67

Pulp

sweetness

0,88 0,69 0,64 0,66 0,91 0,69 0,85 0,72

Pulp acidity 0,66 0,72 0,66

Pulp aroma 0,88 0,68 0,64 0,92 0,68 0,83 0,70

Skin texture 0,71

Skin

astringency

0,89 0,84 0,89

Skin aroma 0,89 0,84 0,95

Skin

bitterness

0,84 0,84 0,92

Seed colour 0,65 0,69 0,70

Seed hardness 0,72 0,65

Seed

bitterness

0,69 0,68 0,92 0,89 0,67 0,90 0,77 0,66

Seed

astringency

0,64 0,64 0,92 0,91 0,71 0,94 0,80 0,78

Seed aroma 0,66 0,66 0,89 0,91 0,67 0,95 0,81 0,73

Pulp maturity 0,67 0,91 0,72 0,92 0,65 0,67 0,71 0,67 0,75 0,86 0,79

Skin maturity 0,71 0,89 0,95 0,92

Seed maturity 0,69 0,66 0,68 0,69 0,72 0,90 0,94 0,95 0,75 0,85 0,82

Berry aroma 0,85 0,83 0,77 0,80 0,81 0,86 0,85 0,93

Berry S

maturity

0,65 0,72 0,70 0,70 0,65 0,66 0,78 0,73 0,79 0,58 0,82 0,93


117

Discriminant analysis on the sensorial characteristics of the grapes at harvest‟s time

highlighted differences between clones examined and all may be well distinct (fig. 60).

The first two canonical functions represent more than 98,9 % of the total variability (tab. 71).


variance %

cumulated Canonical

correlation

1 388,562 80,8 80,8 ,999

2 87,138 18,1 98,9 ,994

3 3,686 ,8 99,7 ,887

4 1,596 ,3 100,0 ,784



points relative to the groups show a very limited dispersion. Five distinct groups appear in the

centroid; above all, the number one, is well detached from the other four.

100,0% of the original grouping are classified correctly, and 100,0% of the cases grouped

cross-validated are reclassified correctly (tab. 72).


harvest‟s time.

118

Thesis

Group expected

Totals 1 2 3 4 5

Crt

oss

- val

idat

ion

a

% 1 100,0 ,0 ,0 ,0 ,0 100,0

2 ,0 100,0 ,0 ,0 ,0 100,0

3 ,0 ,0 100,0 ,0 ,0 100,0

4 ,0 ,0 ,0 100,0 ,0 100,0

5 ,0 ,0 ,0 ,0 100,0 100,0


a.Cross validation is done only for those cases in the analysis in cross validation, each case is




119

3.8 Tecnological and phenolic characteristics of the grapes at

harvest’s time


pH pH n

Sugary content °Brix °Brix

Titratable acidity Titr Ac g/L

Ripening Index* Rip Index n

Bunch weight Bunch wgt g

Berry weight Berry wgt g

Berry seed number Berry seed num n

Berry skin weight Skin wgt/berry g/berry

Berry seed weight Seed wgt/berry g/berry

Anthocyanins/berry Anth/berry mg/berry

Skins % skin %

Skin Anthocyanins Anth skin mg/Kg

Skin Polyphenols Polyph skin mg/Kg

Skin polyphenols percentage % Skin polyph %

Seed polyphenols percentage Polyph seed mg/Kg

Seed percentage Seeds % %

Polyphenols/berry (mg/Kg) Polyph/berry mg/berry

Total Polyphenols Tot Polyph mg/Kg *Ripening Tecnological Index was so calculated °Brix/Titratable acidity


MANOVA was also conducted by examining the parameters (tab. 73) linked to the laboratory

analysis of the grapes (tab. 73) (using the values of three repeated analysis of 17 theses each

year for three years), studying the importance of the variables in function of the chosen source

(tab. 74).

120


Factor Dependent

variable F Sig. Thesis Bunch weight 28579,211 ,000

Berry weight 14,811 ,000

pH 70,538 ,000

°Brix 68,658 ,000

Titratable acidity 28,239 ,000

Ripening index* 30,587 ,000

Berry skin weight 14,126 ,000

Skins % 9,625 ,000

Berry seed number 9,422 ,000

Seed weight/berry 1,302 ,211

Seeds % 1,246 ,247

Anthocyanins skin 5,571 ,000

Polyphenols skin 4,916 ,000

Polyphenols seed 5,540 ,000

Total polyphenols 6,253 ,000

% Skinpolyphenols 3,611 ,000

Anthocyanins/berry 4,118 ,000

Polyphenols/berry 4,635 ,000

Factor Dependent

variable F Sig. Year Bunch weight 7483,430 ,000


pH 407,945 ,000

°Brix 5,791 ,004

Titratable Acidity 80,129 ,000

Ripening Index* 21,621 ,000


Skins % 33,965 ,000


Seed weight/berry ,213 ,808

Seeds % 1,499 ,228








Factor Dependent


* Year Bunch weight 8920,264 ,000


pH 26,873 ,000

°Brix 17,680 ,000


Ripening Index* 12,356 ,000


Skins % 3,276 ,000



Seeds % ,948 ,554








121

All the dependant variables proved to be statistically significant, except two parameters linked

to seeds mean weight/berry and the seed weight percentage compared to the other components

of the berry. Such variables do not change their level of significance if, the factor chosen

during the statistical analysis, is that of the thesis or the year or interaction between the two.

However choosing the year, a non significant statistical variable, anthocyanins /berry is

added.


factor was calculated (tab. 75).

For most of the parameters the greater variability is attributable to the year as shown in the

values for the berry weight, pH, titratable acidity, berry skin weight, skins %, skin

polyphenols, seed polyphenols, total polyphenols and % skin polyphenols. The thesis,

however, shows more variability as concerns bunch weight and °Brix, while the other

parameters show comparable levels of variability attributable to the different source.

High percentage value of error in the two variables linked to the seeds was found.

Variable

%

variability

due to the

thesis

%

variability

due to the

year

%

variability

due to

interaction

thesis/year

%

variability

due to

error Bunch weight 63,53 16,64 19,83 0,00 Berry weight 12,85 76,31 9,97 0,87 pH 13,93 80,56 5,31 0,20 °Brix 73,72 6,22 18,98 1,07 Titratable acidity 23,62 67,02 8,53 0,84 Ripening Index 46,65 32,98 18,85 1,53 Berry skin weight 8,45 87,16 3,80 0,60 Skins % 20,11 70,96 6,84 2,09 Berry seed number 37,32 40,33 18,39 3,96 Seed weight/berry 35,61 5,83 31,20 27,35 Seeds % 26,56 31,94 20,20 21,31 Skin anthocyanins 27,55 48,34 19,17 4,94 Skin polyphenols 12,23 77,25 8,03 2,49 Seed polyphenols 11,11 79,13 7,76 2,01 Total polyphenols 18,89 64,45 13,63 3,02 % Skin polyphenols 6,75 86,80 4,58 1,87 Anthocyanins/berry 39,42 13,07 37,94 9,57


122


non differentiated (tab. 76-78). The significant variables that create the most differentiated

homogenous subsets are those linked to the weight of the berry and the weight of the bunch.

The values obtained from the analysis of the latter form a number of subsets equal to the

number of theses that indicate great variability in the samples tested, the mean weight of the

bunch lies between the minimum value of 167 grams and a maximum of 356 grams, both

belonging to the two thesis of the „Brunello di Montalcino‟ denomination. Moreover the two

thesis mentioned above also differ in the mean weight of the berry, but a lower number of

varied homogenous subsets are formed.

Regarding the pH the values shown are lower in one thesis of the „Chianti Classico‟ and

higher in one of „Montecucco‟, the °Brix grade showing the highest value from one thesis

from the „Brunello di Montalcino‟ production area, and the lowest from the „Chianti Classico‟

area (tab. 76).

The other grapes from the three remaining theses, exception made for those from the

„Morellino di Scansano‟, show medium to high sugar levels, reaching over 28 °Brix.

Moreover the „Brunello di Montalcino‟ denomination shows greater variability in their theses

if compared to „Montecucco‟ (tab. 76).

Regarding titratable acidity, the highest values are found in one thesis in the grapes of the

„Chianti Classico‟ and in one of „Montecucco‟, while the lowest value in the vineyard of the

„Chianti Classico‟ in the „Chianti Colline Pisane‟ (tab. 76), with average values in the

remaining vineyards. Excluding extreme values, within the denominations there is no notable

variability of data.

As for the standard of polyphenols, the mean value is average to low, the overall total is less

present in the „Chianti Classico‟ thesis already highlighted as having the lowest pH values

and sugar levels. The „Brunello di Montalcino‟ thesis, that stands out for its mean value of

3700 mg/Kg of polyphenols expressed as catechin, is that already noted for the highest °Brix

value and for having the lightest berry and bunch (tab. 76). Analysing the total anthocyanins

content two theses emerged as not being notable for the other variables. The lowest value

belongs to a „Brunello di Montalcino‟ thesis while the highest to that of the „Montecucco‟

thesis. Among the variables that are not statistically different seed, weight/berry and

anthocyanins /berry gave to different subsets unlike seeds % (tab. 78).

123

Thesis

Bunch

weight (g)

Berry

weight (g) pH °Brix

Titratable

Acidity (g/L)

Ripen.

Index

Berry

skin

weight (g) Skins %

CCP 1 191,85 b 1,99 bcd 3,60 l 24,61 c 5,01 a 5,04 g 0,39 abc 0,19 ab

MC 1 257,04 i 1,81 b 3,43 gh 25,07 cd 6,96 f 3,66 cd 0,38 ab 0,20 ab

MC 2 239,37 f 2,07cde 3,44 gh 24,82 cd 8,07 h 3,12 bc 0,39 abc 0,19 a

MC 3 335,51 r 2,24 de 3,50 i 23,27 b 6,45 c-f 3,64 cd 0,42 a-d 0,18 a

MC 4 200,37 d 2,00 bcd 3,51 i 25,60 cd 6,33b-f 4,078 def 0,37 a 0,18 a

MC 5 249,52 h 1,87bc 3,67 l 25,05 cd 6,09 b-e 4,11 def 0,37 a 0,19 ab

MC 6 328,79 q 1,97 bcd 3,47 hi 25,89 d 6,40 b-f 4,04 def 0,40 a-d 0,20 ab

MS 1 318,47o 2,09 de 3,30 cd 20,75 a 5,86 bc 3,71 cd 0,39 abc 0,18 a

BM 1 267,11 l 2,33 fg 3,29 cd 25,46 cd 5,82 bc 4,43 ef 0,61 g 0,26 cd

BM 2 242,82 g 1,95bcd 3,34 def 25,31 cd 5,74 abc 4,58 fg 0,46 c-f 0,24 bc

BM 3 195,162 c 1,99 bcd 3,37 ef 25,62 cd 5,88 bc 4,48 ef 0,47 c-f 0,23 bc

BM 4 274,47 n 2,25 efg 3,31 cde 24,97 cd 6,04 bcd 4,16 def 0,48 def 0,21 ab

BM 5 166,08 a 1,607 a 3,38 fg 28,62 e 5,55 ab 5,75 h 0,40 a-d 0,26 cd

BM 6 270,36 m 2,43 g 3,26 bc 23,57 b 6,73 def 3,59 cd 0,52 ef 0,23 abc

BM 7 355,55 s 1,97bcd 3,23 b 25,53 cd 6,78 def 3,89 de 0,55 fg 0,283 d

CC 1 230,09 e 2,09 de 3,28 bcd 20,37 a 6,92 ef 2,97 ab 0,43 a-d 0,20 ab

CC 2 327,45 p 2,14de 3,2 0 a 20,98 a 9,04 i 2,44 a 0,45 a-d 0,21 ab

Table 76. Significant parameters with different and non differentiated subsets. Tukey Test (p=0,05).

Thesis Berry

seed

number

Anthocya-

nins Skin

(mg/Kg)

Polyphe- nols Skin (mg/kg)

Polyphe- nols Seed (mg/kg)

Total

Polyphe- nols (mg/kg)

% Skin

Polyphe- nols

Polyphe- nols /berry

(mg/berry)

CCP 1 2,01 b-e 522abc 1587 a-d 1607 cde 3195 b-e 50 b 6 abcd

MC 1 2,38 def 728d 1703 bcd 1740 e 3443 de 50 b 6 abcd

MC 2 2,48 f 630 bcd 1786 cd 1373 a-e 3160 b-e 43 ab 6 bcd

MC 3 2,56 f 661bcd 1292 a 1351 a-e 2644 ab 49 b 5 abc

MC 4 2,31 b-f 615 bcd 1547 abc 1465 b-e 3013 a-d 48 b 6 abc

MC 5 2,42 ef 594 bcd 1734 bcd 1440 a-e 3174 b-e 44 ab 5 abc

MC 6 2,20 b-f 580 bcd 1596 a-d 168 de 3282 cde 51 b 6 bcd

MS 1 2,32 def 513 abc 1493 abc 1360 a-e 2853 a-d 47 b 6 abc

BM 1 2,60 f 359 a 1432 abc 1408 a-e 2841 a-d 49 b 6 bcd

BM 2 2,16 b-f 569 bcd 1581 a-d 1260 a-d 2842 a-d 44 ab 5 ab

BM 3 1,56 a 473 ab 1518 abc 1289 a-d 2807 abc 45 b 6 abc

BM 4 1,87 abc 521 abc 1382 ab 1193 abc 2575 ab 46 b 5 abc

BM 5 1,93 a-d 671 cd 1949 d 1780 e 3729 e 47 b 5 abc

BM 6 2,19 b-f 487 abc 1504 abc 1482 -e 2987 a-d 47 b 7 d

BM 7 1,78 ab 653 bcd 1781 cd 1028 a 2810 abc 35 a 5 a

CC 1 2,15 bcdef 624 bcd 1753 bcd 1440 a-e 3194 b-e 45 b 6 bcd

CC 2 2,37 def 508 bc 1430 abc 1092 ab 2522 a 42 ab 5 ab

Table 77. Significant parameters with non differentiated subsets. Tukey Test (p=0,05).

124

Thesis

Seed

weight/berry

(g/berry)

Seeds %

Anthocyanins

/berry

(mg/berry)

CCP 1 0,09 ab 0,05 a 1,01 a-d

MC 1 0,10 ab 0,05 a 1,31 cd

MC 2 0,10 ab 0,05 a 1,36 d

MC 3 0,10 ab 0,04 a 1,33 d

MC 4 0,12 ab 0,06 a 1,23 bcd

MC 5 0,08 ab 0,04 a 1,10 a-d

MC 6 0,11 ab 0,06 a 1,13 a-d

MS 1 0,09 ab 0,04 a 1,06 a-d

BM 1 0,11 ab 0,05 a 0,87 a

BM 2 0,09 ab 0,04 a 1,09 a-d

BM 3 0,06 a 0,03 a 0,93 ab

BM 4 0,08 ab 0,03 a 1,15 a-d

BM 5 0,07 ab 0,05 a 0,97 abc

BM 6 0,10 ab 0,04 a 1,20 a-d

BM 7 0,07 ab 0,03 a 1,26 bcd

CC 1 0,08 ab 0,04 a 1,26 bcd

CC 2 0,22 b 0,10 a 1,08 a-d

Table 78. Non significant parameters with non differentiated subsets. Tukey Test (p=0,05).

From the multivariate analysis factor choice, the year and the Tukey statistic test it appears

that most of the parameters tested originate differentiated subsets, exception made for the

variables seed weight/berry, anthocyanins/berry and seed%. In bunch weight, berry weight,

pH, berry skin weight, skins% and seed polyphenols, instead, it is the variables that create

well differentiated subsets that indicate a great variability of data in the three year period

studied. The year 2009 shows the highest parameters that influence the technological ripeness

of the grapes while in the following two years higher quantities of polyphenols are registered

(tab.79).

Variable 2009 2010 2011

Bunch weight (g) 274,97 c 251,167 a 259,16 b

Berry weight (g) 2,26 c 1,85 a 2,03 b

pH 3,53c 3,28 a 3,35 b

°Brix 24,60 b 24,58 b 24,15 a

Titratable acidity (g/L) 6,08 a 7,24 b 6,04 a

Ripening Index 4,21 b 3,68 a 4,06 b

Berry skin weight (g) 0,54 c 0,35 a 0,44 b

Skins % 0,24 c 0,19 a 0,22 b

Berry seed number 2,34 b 2,16 a 2,07 a

125

Variable 2009 2010 2011

Seed weight/berry (g/berry) 0,10 a 0,11 a 0,10 a

Seeds % 0,04 a 0,06 a 0,05 a

Skin anthocyanins (mg/Kg) 513 a 609 b 591 b

Skin polyphenols (mg/Kg) 1528 a 1452 a 1797 b

Seed polyphenols (mg/Kg) 1193 a 1662 c 1380 b

Total polyphenols (mg/Kg) 2721 a 3115 b 3177 b

% Skin polyphenols 43,08 a 52,86 b 43,37 a

Anthocyanins/berry(mg/Kg) 1,15 a 1,10 a 1,17 a

Polyphenols/berry (mg/Kg) 6,08 b 5,64 a 6,32 b


horizontal.


and calculated in four new complex variables (components) so as to represent 92,53% of the

total variability of the macro structural characteristics of the berries at harvest (tab. 80). The


determining the characteristics of technological ripeness and phenol ripeness of the berries at

harvest. The first component is tied to the weight of the bunch, to the number and grade of the

polyphenols in the seeds, to the percentage and weight of the skins, the second, instead, to the

weight of the berry and to the polyphenols per berry. The total polyphenols, of the skins and

the polyphenols per berry, anthocyanins of the skins and per berry are the variables that

characterize the third component calculated by factorial analysis while the fourth by the seed

as a percentage value compared to the other components of the berry and as weight expressed

in grams (tab. 81).

Component


Total

%

variability

%

cumulated

1 5,62 31,23 31,23

2 4,53 25,15 56,39

3 4,20 23,34 79,73

4 2,30 12,80 92,53


126

Variable Component

1 2 3 4 Seed Polyphenols ,955

Bunch weight ,768 -,605

% Skin polyphenols ,754 -,625

Berry seed number ,735 ,459

Total polyphenols ,767

Skin anthocyanins -,442 ,779

Seeds % -,493 ,762

Anthocyanins/berry ,957

Titratable acidity -,927

Seed weight/berry ,976


Polyphenols /berry ,790 ,507

°Brix ,501 ,582

pH -,649 ,441 -,490

Berry weight -,662 ,729

Berry skin weight -,848 ,420

Skins % -,863





In particular the first component is greatly linked to the polyphenol content of the skin, to the

total polyphenols, to the mean weight of the berry and to that of the seed in terms of the total

weight of the berry, to the total of pH and acidity. The second on the other hand, to the

polyphenols of the seeds and to the grade of polyphenols per berry (tab. 83).



negatively. The Ripening Index is indeed positively correlated with the sugar level and with

the grade of polyphenols in the berry and skins and negatively with the parameters tied to the

characteristics of the seed. Moreover the first component is positively correlated with all the

variables examined, while the second is correlated only in part, in fact it is tied negatively to

the acidity measure worthy of note, to the polyphenols present in the skins and seeds and to

the average weight of the seeds and of the bunch.

127

Component


Total

%

variance

%

cumulated 1 6,50 36,13 36,13

2 6,07 33,72 69,85



Variable Component

1 2 Skin polyphenols ,829


Berry weight ,760 -,577


% Skin polyphenols ,645 ,712

Polyphenols /berry ,542 ,715

Skin anthocyanins ,541 -,630

Anthocyanins/berry -,606

Ripening Index -,548

Seed Polyphenols ,888

Seeds % -,317



Polyphenols/berry ,929

pH -,762 ,542

°Brix -,834

Berry seed number -,895

Skins % -,913


order of importance excluding the <0,4 coefficients as of little relevance .

128


Analysing the multiple linear regression in the four new components and the ripeness index of

the berry, it is possible to note that there is a significant correlation and that the total ripeness

of the berry obtained experimentally is linearly correlated in a significant way (r2 = 0,822) to

that estimated in (tab. 84). Therefore, the total ripeness index of the berry can be expressed in

the following way (tab. 85).

Ripening Index =3,241-F1*0,124+F2*0,729+F3*0,236-F4*0,078

From the table of coefficient values for the estimation of the total ripeness of the berry, it can

be concluded that the model used is of significant importance, that the Ripening Index

variable, is more relevant with the last three constants rather than with the first (tab. 84).


Standard deviation Estimate‟s error

1 ,907 ,822 ,818 ,470

Table 84. Statistic model of relation between the dependant variable of total ripeness of the berry and

the four new components obtained from factorial analysis.

129

Coefficients B

Standard

deviation


F 1 -,124 ,052 ,018

F 2 ,729 ,031 ,000

F 3 ,236 ,048 ,000

F 4 -,078 ,013 ,000

Table 85. Coefficient values for the estimation of the total maturity estimated of the berry,


Calculating the relation between the index ripeness expressed and that estimated statistically it

can be seen graphically how the ripeness expressed by the panel and that determined

statistically are overlapping (fig. 62).



with a probability <0,00001 are reported. and the characters in bold indicate the negativity of

the value indicated.

In most cases there are values of Pearson's correlation positive. However the parameters

connected with the weight of the bunch, and grape skins, are negatively correlated with levels

of total polyphenols, and polyphenols found in the skins and seeds (tab. 86).

130

Bunch

wgt

Berry

wgt pH °Brix

Titr

ac

Rip

Index

Berry

skin

wgt

Skins

%

Berry

seed

num

Seed

wgt/

berry

Seeds

%

Skin

anth

Skin

polyph

Seed

polyph

Tot

polyph

%

Skin

polyph

Anth/

berry

Polyph/

berry F 1 F 2 F 3 F 4

Bunch weight 0,42 0,36 0,37 0,40

Berry weight 0,33 0,29 0,70 0,43 0,46 0,46 0,50 0,62 0,41 0,47 0,36

pH 0,39 0,44

°Brix 0,42 0,33 0,57 0,51 0,34

Titr Acidity 0,39 0,82 0,34 0,81

Rip Index 0,57 0,82 0,37 0,87

Berry skin

weight

0,70 0,70 0,33 0,50 0,46 0,34 0,53

Skins % 0,70 0,33

Berry seed

number

0,43

Seed

weight/berry

0,98 0,58 0,99

Seeds % 0,98 0,49 0,27 0,96

Anthocyanins

skin 0,46 0,33 0,42 0,38 0,75 0,72

Polyphenols

skin 0,46 0,42 0,72 0,41 0,72

Polyphenols

seed 0,36 0,50 0,50 0,83 0,78 0,39 0,74

Total

Polyphenols 0,37 0,62 0,46 0,38 0,72 0,83 0,44 0,60 0,47

% Skin

polyphenols 0,34 0,33 0,41 0,78 0,59 0,38

Anthocyanins

berry

0,75 0,54

Polyphenols/

berry

0,41 0,39 0,44

F 1 0,47 0,34 0,37 0,53 0,58 0,49 0,74 0,60 0,59 0,37 0,62

F 2 0,40 0,44 0,51 0,81 0,87 0,37

F 3 0,36 0,34 0,72 0,72 0,47 0,38 0,54

F 4 0,99 0,96 0,62

Table 86. Pearson‟s correlations. Bold numbers are preceded by the minus sign. Legend abbreviations is on the previous pages.

131

The statistical analysis was also used to investigate the features and possible statistically

significant differences of the grapes coming from the theses as part of the same Denomination

of Origin. The theses, divided in „Montecucco‟, were then subjected to MANOVA, factorial,

discriminating, linear regression and correlation analysis.

Among the theses coming from the area of „Montalcino‟ was also studied the case of „Col

D'Orcia‟ estate in order to study in detail the clone effect grown in the same site of

cultivation.

3.8.1 ‘Montecucco’ area

As results from the MANOVA, statistically significant parameters do not remain the same if

the factor chosen during the statistical analysis is changed. Berry‟s and bunch‟s weight, pH,

titratable acidity, ripening index, the percentage of polyphenols of skins and polyphenols per

berry represent the dependent variables that remain no statistically different. The parameters

related to polyphenols become no significant when the factor is represented by year or

interaction year by thesis (tab. 87).

Factor Dependent

variable F Sig. Factor Dependent

variable F Sig. Thesis Bunch weight 13303,90 ,000 Year Bunch weight 11899,079 ,000

Berry weight 14,306 ,000 Berry weight 33,938 ,000

pH 10,549 ,000 pH 74,902 ,000

°Brix 16,301 ,000 °Brix 2,184 ,131

Titratable acidity 17,630 ,000 Titratable acidity 21,076 ,000

Ripening Index 19,012 ,000 Ripening Index 12,819 ,000

Berry skin weight ,864 ,517 Berry skin weight 52,663 ,000

Skins % 1,009 ,430 Skins % 30,503 ,000

Berry seed number ,598 ,702 Berry seed number 4,494 ,020

Seed weight/berry 1,650 ,179 Seed weight/berry ,591 ,560

Seeds % 1,253 ,311 Seeds % ,577 ,568

Skin anthocyanins 2,920 ,030 Skin anthocyanins 2,207 ,128

Skin polyphenols 2,260 ,075 Skin polyphenols 17,000 ,000

Seed polyphenols 2,791 ,035 Seed polyphenols 9,198 ,001

Total polyphenols 1,546 ,207 Total polyphenols ,395 ,677

% Skin polyphenols 5,330 ,001 % Skin polyphenols 41,904 ,000

Anthocyans /

berry 3,166 ,021

Anthocyans /berry

,690 ,509

Polyphenols/berry 2,766 ,037 Polyphenols/berry 10,682 ,000

132

Factor Dependent

variable F Sig. Thesis * Year Bunch weight 3622,812 ,000


pH 11,079 ,000

°Brix 17,868 ,000


Ripening Index 4,018 ,003


Skins % 2,991 ,017



Seeds % 1,610 ,172

Skin anthocyanins 3,775 ,005

Skin polyphenols 1,264 ,302

Seedpolyphenols 3,796 ,005


% Skin polyphenols 1,727 ,142

Anthocyans /berry 3,447 ,008


Table 87 a, b, c. Test of effects between subjects (p=<0,05).



year; the thesis, however, shows more variability as concerns ripening index, the other

parameters show comparable levels of variability attributable to the different factor.

133

Variable

%

variability

due to the

thesis

%

variability

due to the

year

%

variability

due to

interaction

thesis/year

%

variability

due to

error

Bunch weight 46,15 41,28 12,57 0,00 Berry weight 22,91 54,35 21,14 1,60 pH 10,82 76,80 11,36 1,03 °Brix 43,64 5,85 47,83 2,68 Titratable acidity 40,16 48,02 9,54 2,28 Ripening Index 51,60 34,79 10,90 2,71 Berry skin weight 1,49 90,61 6,18 1,72 Skins % 2,84 85,92 8,42 2,82 Berry seed number 6,21 46,68 36,73 10,39 Seed weight/berry 37,33 13,38 26,65 22,63 Seeds % 28,23 12,99 36,25 22,52 Skin anthocyanins 29,49 22,29 38,12 10,10 Skin polyphenols 10,50 78,98 5,87 4,65 Seedpolyphenols 16,63 54,80 22,62 5,96 Total polyphenols 25,20 6,45 52,05 16,30 % Skin polyphenols 10,67 83,87 3,46 2,00 Anthocyans /berry 38,13 8,31 41,51 12,04 Polyphenols/berry 16,24 62,71 15,18 5,87



non differentiated (tab 89-91). The significant variables that create the most differentiated

homogenous subsets are those linked to the mean weight of the bunch, to the sugary and acids

content, to the ripening index, and to the percentage of polyphenols contained in the skin.

The values obtained from the analysis of the mean weight of the bunch form a number of

subsets equal to the number of thesis that indicate great variability in the samples tested,

indeed this variable lies between the minimum value of 200 grams and a maximum of 300

grams. The sugar concentration is medium high; the thesis with the lowest value shows about

24° Brix and the one with the highest value is reaching over 28 °Brix: this latter thesis is the

same already mentioned for the heaviest bunch which also differs from the others for the

greatest total polyphenols content and the lowest in anthocyanins.

The berries of the theses show good values of total acidity, this variable in one case has

reached 8 g/L (tab. 90-91).

Between not statistically different variables, only the variable seed weight/berry originates

different subsets (tab. 91).

134

Thesis

Bunch

weight (g)

Berry

weight (g) pH °Brix

Titratable

Acidity (g/L)

Ripen.

Index

Skin

Anthocya-

nins

(mg/kg)

Seed

Polyphe-

nols (mg/Kg)

MC 1 255,47 d 1,75a 3,42 ab 25,07 b 6,92 a 3,68 bc 737b 1817 ab

MC 2 239,37 c 2,07 b 3,43 ab 24,82 ab 8,07 b 3,12 a 661 ab 1374 a

MC 3 290,30 e 2,09 b 3,41 a 23,97 a 6,98 a 3,45 ab 694 ab 1685 ab

MC 4 200,37 a 2,00 b 3,51 bc 25,6 bc 6,33 a 4,07 cd 615ab 1465 ab

MC 5 219,37 b 1,67 a 3,55 c 26,25 c 6,32 a 4,16 d 602 ab 1698 ab

MC 6 305,71 f 2,02 b 3,56 c 27,6 d 6,50 a 4,25 d 498 a 1906 b

Table 89. Significant parameters with differentiated subsets. Tukey Test (p=0,05).


Thesis

Berry

skin

weight Skins

%

Berry

seed

number Seeds

% % Skin

Polyphenols

Total

Polyphenols

(mg/kg)

Seed

weight/

berry

MC 1 0,36 a 0,20 a 2,42 a 0,06 a 1735,10 a 3552 a 0,10 ab

MC 2 0,39 a 0,19 a 2,49 a 0,05 a 1786,98 a 3161 a 0,10 ab

MC 3 0,36 a 0,17 a 2,42 a 0,04 a 1464,91 a 3149 a 0,09 ab

MC 4 0,37 a 0,18 a 2,31 a 0,06 a 1547,87 a 3013 a 0,12 b

MC 5 0,30 a 0,18 a 2,23 a 0,04 a 1769,04 a 3467 a 0,07 a

MC 6 0,38 a 0,19 a 2,42 a 0,05 a 1489,25 a 3395 a 0,10 ab

Table 91. Non significant parameters with different and non differentiated subsets. Tukey Test

(p=0,05).

The year appears the most of the parameters tested originate differentiated subsets, exception

made for total polyphenols and for variables related to anthocyanins and to seeds. The year

2009 shows the highest parameters that influence the technological ripeness of the grapes at

the harvest while in the following two years higher quantities of polyphenols are registered

(tab. 92).

Thesis

Polyphenols/ berry (mg/berry)

% Skin

Polyphenols

Anthocyanins/

berry

(mg/berry)

MC 1 6,14 ab 50,86 bc 1,29 a

MC 2 6,50 ab 43,19 a 1,36 a

MC 3 6,47 ab 53,04 bc 1,36 a

MC 4 5,99 ab 48,27 ab 1,23 a

MC 5 5,75 a 48,94 ab 1,00 a

MC 6 6,86 b 56,21 c 1,00 a

135

Variable 2009 2010 2011 Bunch weight 229,96 a 237,31 b 272,84 c

Berry weight 2,12 c 1,78 a 1,99 b

pH 3,67 c 3,44 b 3,38 a

°Brix 26,20 b 25,51 a 24,92 a

Titratable acidity 6,86 b 7,36 c 6,38 a

Ripening Index 3,88 b 3,53 a 3,94b

Berry skin weight 0,47 c 0,28 a 0,38 b

Skins % 0,22 c 0,15 a 0,19 b

Berry seed number 2,65 b 2,26 a 2,34 a

Seed weight/berry 0,11 a 0,09 a 0,09 a

Seeds % 0,05 a 0,05 a 0,05 a

Skin anthocyanins 590,39 a 684,46 a 622,34 a

Skin polyphenols 1719,39 b 1399,44 a 1875,36 b

Seed polyphenols 1494,45 a 1873,56 b 1443,98 a

Total polyphenols 3213,84 a 3272,00 a 3319,34 a

% Skin polyphenols 45,49 a 57,14 b 43,35 a

Anthocyans /berry 1,26 a 1,20 a 1,20 a

Polyphenols/berry 6,79 b 5,74 a 6,52 b


horizontal.

By using the factorial statistic analysis it was possible to put together all the variables noted


total variability of the technological characteristics of the berries at harvest (tab. 93). The


determining the characteristics of technological ripeness and phenol resources of the berries at

harvest.

There are only three values of the coefficients less than 0.4, including two linked to the

titratable acidity, and then all remaining variables have influence for the purpose of analysis.

The first component is tied to the most of the parameters studied except for the value of total

acidity, mean weight of berry and of anthocyanins present in the skins; these values are

associated with the second component. The last component is characterized, however, to pH

and polyphenols content of skins (tab. 94).

136

Component


Total % variability % cumulated 1 8,91 49,51 49,51

2 5,28 29,35 78,87

3 2,99 16,63 95,49


Variable Component

1 2 3 Polyphenols/berry ,986 ,094



°Brix ,816 ,483

Skin polyphenols ,811 ,553

Skins % ,788 ,580 Berry skin weight ,721 ,496 ,463

Seeds % ,706 ,599 Seed weight/berry ,693 ,657 Berry seed number ,644 ,472 ,601

pH -,896

Titratable acidity -,951 % Skin polyphenols -,699

Berry weight -,807 -,457

Skin anthocyanins -,740 -,548

Anthocyans /berry -,867 -,463

Bunch weight -,867 -,463


below 0,4 are not included because unimportant.



In particular the first component is greatly linked to the variables already mentioned for the

description of the first component; the second is composed of the sum of the second and third

component of the previous analysis (tab. 96).



negatively. The Ripening Index is indeed positively correlated with the sugary content, the

pH, the seeds‟s and skins‟s weight, and with the grade of polyphenols in the berry and skins.

137

On the contrary, it is negatively correlated with the mean weight of the berry and of the bunch

and the titratable acidity.

The first and the second component are negatively correlated with the antochyanins and the

mean weight of the berry and of the bunch.

Component


Total

%

variance

%

cumulated 1 9,55 53,04 53,04

2 6,35 35,29 88,33



Variable Component

1 2 Polyphenols/berry ,993



°Brix ,847 ,524

Skin polyphenols ,830 ,476

Skins % ,821 ,556


Seeds % ,749 ,651



pH ,695 ,674

% Skin polyphenols -,803

Titratable acidity -,697

Berry weight


Anthocyans /berry -,424 -,893

Bunch weight -,889 -,437



138


Analysing the multiple linear regression in the three new components and the ripeness index

of the berry, it is possible to note that there is a significant correlation and that the total

ripeness of the berry obtained experimentally is linearly correlated in a significant way (r2 =

0,665) to that estimated in (tab. 97). Therefore, the total ripeness index of the berry can be

expressed in the following way (tab. 98).

Ripening Index = 4,006+F1*0,99+F2*0,160+F3*0,152


be concluded that the model used is of significant importance, though the Ripening Index

variable, is more relevant with the last constant rather than with the first two (tab. 98).

139


Standard

deviation Estimate’s error

1 ,815 ,665 ,640 ,351


the three new components obtained from factorial analysis.

Coefficients B Standard deviation


F1 ,099 ,103 ,340

F2 ,160 ,055 ,006

F3 ,152 ,031 ,000

Table 98. Coefficient values for the estimation of the total ripeness estimated of the berry,



with a probability <0,00001 are reported and the characters in bold indicate the negativity of


There are a few significant Pearson‟s correlations and they are expressed in like way from

positive and negative values.

Bunch weight, pH, berry seed number, polyphenols/berry, are the parameters that don‟t present

significant Pearson‟s correlations (tab. 99).

140

Variable Berry

weight

(g) °Brix

Titr.

Ac.

(g/L)

Rip

Index

Berry

skin

weight

(g)

Seed

weight/

berry

(g)

Seeds

%

Skin

anthocy.

(mg/Kg)

Skin

polyph.

(mg/Kg)

Seed

polyph.

(mg/Kg)

Total

polyph.

(mg/Kg)

%

Skin

polyph.

Anthocy./

berry

(mg/berry)

Berry weight

(g)

0,68 0,66

Titratable acidity

(g/L)

0,88

Ripening Index 0,57 0,88

Berry skin weight (g) 0,68

Skins % 0,88

Seed weight /berry

(g/berry)

0,88

Seeds % 0,88

Skin anthocyanins

(mg/Kg)

0,86

Skin polyphenols

(mg/Kg)

0,64

Seed polyphenols

(mg/Kg) 0,66 0,78 0,79

Total polyphenols

(mg/Kg)

0,78

% Skin polyphenols 0,64 0,79

Anthocyanins/berry

(mg/berry)

0,86


141

Discriminant analysis on the technological and phenolic characteristics of the grapes at harvest‟s time

highlighted differences between areas examined and some of which may be distinct (fig. 64).

The first two canonical functions represent more than 77,0 % of the total variability (tab.

100).


variance %

cumulated Canonical

correlation

1 12,212 57,8 57,8 ,961

2 4,083 19,3 77,1 ,896

3 3,029 14,3 91,5 ,867

4 1,457 6,9 98,4 ,770

5 ,346 1,6 100,0 ,507



points relative to the groups show a limited dispersion. Two distinct groups appear in the

centroid: the first includes the thesis coming from a different a vineyard of a different farm from

the other five theses and that is the only one that well differs from the other, because well detached.

The second the residual theses: the number two is the most distinguishable while some points

linked to number four and five are superimposed.

The 95,5% of the original grouping are classified correctly, while 68,2% of the cases grouped

cross-validated are reclassified correctly (tab. 101).

142

Figure 64. Centroids obtained from the cluster analysis of the characteristics of the grapes at harvest‟s

time.

Thesis

Group expected

Totals 1 2 3 4 5 6 Cross-validation

a % 1 33,3 ,0 50,0 16,7 ,0 ,0 100,0

2 ,0 66,7 16,7 ,0 ,0 16,7 100,0

3 37,5 12,5 50,0 ,0 ,0 ,0 100,0

4 11,1 ,0 11,1 66,7 11,1 ,0 100,0

5 ,0 ,0 ,0 11,1 88,9 ,0 100,0

6 ,0 ,0 ,0 ,0 ,0 100,0 100,0

Table 101. Classification results. a. Cross validation is done only for those cases in the analysis in cross validation, each case is



143


As results from the multivariance analysis the dependant variables proved to be statistically

significant.

If the thesis is chosen as factor, titratable acidity and percentage of seed‟s polyphenols

become non significant statistical variables; instead choosing the year ripening index, the

percentage of skins and the anthocyanins‟s content, become non significant statistical

variables.

Seed‟s percentage, polyphenols level and skin‟s percentage polyphenols are the parameters

that become non significant when the factor is represented by year by thesis (tab. 102).

Factor Dependent

variable F Sig. Factor Dependent

variable F Sig. Thesis Bunch weight 19832,816 ,000 Year Bunch weight 10903,491 ,000

Berry weight 62,925 ,000 Berry weight 77,761 ,000

pH 11,196 ,000 pH 466,432 ,000

°Brix 22,696 ,000 °Brix 7,985 ,002

Titratable acidity

1,904 ,136

Titratable acidity

19,359 ,000

Ripening Index 8,226 ,000 Ripening Index ,040 ,961

Berry skin weight

29,807 ,000

Berry skin weight

42,400 ,000

Skins % 2,925 ,037 Skins % 1,374 ,269

Berry seed number

21,754 ,000

Berry seed number

6,252 ,005

Seed weight/ berry

11,557 ,000

Seed weight/ berry

5,403 ,010

Seeds % 4,249 ,008 Seeds % 5,840 ,007

Skin anthocyanins

42,798 ,000

Skin anthocyanins

14,454 ,000

Skin polyphenols

13,246 ,000

Skin polyphenols

6,225 ,005

Seed polyphenols

7,521 ,000

Seed polyphenols

3,669 ,038

Total polyphenols

20,314 ,000

Total polyphenols

4,967 ,014

% Skin polyphenols

,941 ,454

%Skin polyphenols

4,243 ,024

Anthocyans

/berry 14,924 ,000

Anthocyans / berry

1,807 ,182

Polyphenols/ berry

4,892 ,004

Polyphenols/ berry

9,494 ,001

144

Factor Dipendent variable F Sig. Thesis *

Year Bunch weight 3184,300 ,000


pH 11,801 ,000

°Brix 6,298 ,000


Ripening Index 13,840 ,000


Skins % 3,073 ,012



Seeds % ,675 ,709

Skin anthocyanins 26,345 ,000

Skin polyphenols 4,834 ,001

Seed polyphenols 1,217 ,323


% Skin polyphenols ,607 ,765

Anthocyans /berry 25,144 ,000




factor was calculated (tab. 103).

For most of the parameters the variability is attributable to the thesis; the year, however,

shows more variability as concerns berry‟s and seed‟s weight, pH, seed‟s and polyphenols‟s

percentage and polyphenolic content per berry.

Titratable acidity, ripening index and anthocyanins content per berry are the parameters that

show more variability when the factor is represented by year interaction by thesis (tab. 103).

The variability quota attributed to the different factor was calculated using the data previously

obtained.

145

Variable

%

variability

due to

the thesis

%

variability

due to

the year

%

variability

due to

interaction

thesis/year

%

variability

due to

error

Bunch weight 58,47 32,14 9,39 0,00 Berry weight 33,66 41,60 24,21 0,53 pH 2,28 95,11 2,41 0,20 °Brix 59,76 21,02 16,58 2,63 Titratable acidity 3,28 33,31 61,69 1,72 Ripening Index 35,60 0,17 59,90 4,33 Berry skin weight 31,33 44,56 23,07 1,05 Skins % 34,94 16,41 36,71 11,94 Berry seed number 61,41 17,65 18,11 2,82 Seed weight/berry 49,18 22,99 23,57 4,26 Seeds % 36,11 49,64 5,74 8,50 Skin anthocyanins 50,59 17,09 31,14 1,18 Skin polyphenols 52,35 24,60 19,10 3,95 Seed polyphenols 56,10 27,37 9,08 7,46 Total polyphenols 65,32 15,97 15,49 3,22 % Skin polyphenols 13,86 62,48 8,94 14,72 Anthocyans /berry 34,81 4,21 58,64 2,33 Polyphenols/berry 20,02 38,85 37,03 4,09



non differentiated (tab. 104-106).

The significant variables create differentiated homogenous subsets except for titratable

acidity and seed‟s polyphenols expressed by percentage (tab. 104-105).

The values obtained from the analysis of bunch‟s weight form a number of subsets equal to

the number of theses that indicate great variability in the samples tested, the mean weight of

the bunch lies between a minimum value of 167 grams and a maximum of 274 grams. Berry‟s

weight, instead, creates less homogenous subsets.

The thesis BM5, stands out from others ones, for the highest value of bunch‟s weight, of

sugary content, and of polyphenols‟s supply; besides, this thesis shows the lowest value of pH

(tab. 104-106).

146

Thesis Bunch

weight

(g)

Berry

weight

(g)

pH °Brix Ripening

Index

Berry

skin

weight

(g)

Skins

%

Berry

seed

number

Seed

weight/

berry

(g/berry)

BM 1 267,11 d 2,33 c 3,29 a 25,46 a 4,43 a 0,61 c 0,26 b 2,60 c 0,11 c

BM 2 242,82 c 1,95 b 3,33 ab 25,31 a 4,58 a 0,46 b 0,24 ab 2,16 b 0,09 b

BM 3 195,16 b 1,99 b 3,36 bc 25,62 a 4,48 a 0,47 b 0,23 ab 1,56 a 0,06 a

BM 4 274,46 e 2,25 c 3,30 a 24,97 a 4,16 a 0,48 b 0,21 a 1,87 b 0,08 ab

BM 5 166,08 a 1,60 a 3,38 c 28,62 b 5,74 b 0,40 a 0,26 b 1,93 b 0,07 ab

Table 104. Significant parameters with differentiated subsets. Tukey Test (p=0,05).

Thesis Seeds

%

Skin

anthocyanins

(mg/kg)

Seed

polyphenols

(mg/kg)

Total

polyphenols

(mg/kg)

Anthocyans /

berry

(mg/berry)

Polyphenols/

berry

(mg/berry)

Skin

polyphenols

(mg/kg)

BM 1 0,05 b 359 a 1409 a 2841 a 0,87 a 6,51 b 1433 a

BM 2 0,04 b 570 c 1261 a 2842 a 1,09 b 5,48a 1581 a

BM 3 0,03 a 473 b 1289 a 2808 a 0,93 a 5,58 a 1519 a

BM 4 0,03 ab 521 bc 1193 a 2575 a 1,15 b 5,76 a 1382 a

BM 5 0,04 b 671d 1780 b 3729 b 0,97 a 5,70 a 1949 a


Thesis Titratable

acidity (g/L)

%Skin

polyphenols

BM 1 5,81 a 49,52 a

BM 2 5,74 a 44,53 a

BM 3 5,88 a 45,94 a

BM 4 6,04 a 46,02 a

BM 5 5,55 a 47,58 a

Table 106. Non significant parameters with different and non differentiated subsets. Tukey Test

(p=0,05).

From the multivariance analysis where factor choice is the year and the Tukey statistic test, it

appears that most of the parameters tested originate differentiated subsets, exception made for

the variables ripening index, anthocyanins‟s content per berry and percentage of skins.

The years 2009 and 2010 show the highest parameters that influence the technological

ripeness of the grapes; the highest quantities of polyphenols, instead are registered in 2011

(tab. 107).

In the year 2009 are highlighted the parameters that influence the technological ripeness of

the grapes while in the following two years higher quantities of polyphenols are registered; in

2010 stood out the parameters that influence the phenolic richness (tab. 107).

147

Variable 2009 2010 2011 Bunch weight 260,24 c 214,36 b 212,773a

Berry weight 2,29 c 1,81 a 1,97 b

pH 3,45 c 3,12 a 3,42 b

°Brix 25,44 a 26,76 b 25,80 a

Titratable acidity 5,58 a 6,31 b 5,52 a

Ripening Index 4,65 a 4,67 a 4,72 a

Berry skin weight 0,56 c 0,42 a 0,48 b

Skins % 0,24 a 0,23 a 0,25 a

Berry seed number 2,18 b 2,03 ab 1,86 a

Seed weight/berry 0,08 ab 0,07 a 0,10 b

Seeds % 0,03 a 0,04 ab 0,05 b

Skin anthocyanins 466 a 570 c 519 b

Skin polyphenols 1474 a 1538 a 1705 b

Seed polyphenols 1288 a 1529 b 1341 ab

Total polyphenols 2763 a 3068 b 3047b

% Skin polyphenols 46,91 ab 49,74 b 43,50 a

Anthocyans /berry 1,04 a 0,98 a 0,98 a

Polyphenols/berry 6,26 b 5,39 a 5,76 a

Table 107. Mean separation by multiple range test (Tukey) the comparison is among data shown in

horizontal.



total variability of the technological characteristics of the berries at harvest (tab. 108). The


determining the characteristics of technological ripeness and phenol richness of the berries at

harvest‟s time.

Only three values present coefficients of less than 0,4 and two of them are linked to the

titratable acidity and so all the remaining operate in a more reliable way in determining the

characteristics of technological ripeness and phenol richness of the theses.

The first component is tied to the sugar content and to the seeds (expressed in mean number

per berry and in berry‟s total weight); the second, instead to the pH, to the titratable acidity, to

the bunch‟s weight and to the anthocyanins‟s content.

The seed‟s and skin‟s percentage are the variables that characterize the third component

calculated by factorial analysis while the fourth is marked by the polyphenols located in the

seeds and in the skins (tab. 109).

148

Component


Total %

variability %

cumulated 1 6,61 36,74 36,74

2 6,19 34,43 71,17

3 2,29 12,75 83,92

4 2,21 12,27 96,19


Component

1 2 3 4 Seed weight/berry ,912


Polyphenols/berry ,843 ,419

Berry weight ,827 ,467


Anthocyans /berry ,738 -,492


% Skin polyphenols ,894


pH ,906

Seeds % -,442 ,852

Skins % ,834



Total polyphenols -,614 ,516

Skin polyphenols -,785 ,462

°Brix -,895



Extracting only two components from all the data collected it is possible to explain 78,42 %

of the total variability (tab. 110).

In particular the first component is greatly linked to the most of the variables analysed in the

study; the second, on the other hand, to only seed‟s anthocyanins and poliphenols content.

(tab. 111).



negatively. The Ripening Index is indeed positively correlated with the seed‟s polyphenols,

149

and with skin‟s anthocyanins and negatively with the pH, the titratable acidity and bunch‟s

mean weight.

The first and the second component are negatively correlated with the sugar and the total

polyphenols content and with the percentage of skins and seeds.

Moreover the first component is negatively correlated with seed‟s anthocyanins and

polyphenols, while the second is negatively correlated with the acidity and the bunch‟s

weight.

Component


Totale % variance % cumulated

1 10,380 57,665 57,665

2 3,737 20,762 78,427

Table 110. Results of the factorial analysis extracting only the first two components: method of the

main components.

Component

1 2 Berry weight ,971


Polyphenols/berry ,921




pH ,789 -,567

Seed weight/berry ,775

Anthocyans /berry ,878

% Skin polyphenols ,669

Skins % -,643

Seeds % -,545

Seed polyphenols -,616

Skin polyphenols -,783 -,517

°Brix -,839

Total polyphenols -,862

Skin anthocyanins -,868 ,450


order of importance excluding the <0,4 coefficients as of little relevance .

150


Analysing the multiple linear regression in the four new components and the ripeness index of

the berry, it is possible to note that there is a non significant correlation and that the total

ripeness of the berry obtained experimentally is slightly linearly correlated (r2 = 0,344) to that

estimated in (tab. 112).

Model R R-square R-square

correct

Standard Deviation Estimate’s error

1 ,586 ,344 ,278 1,09


the four new components obtained from factorial analysis.

151

Therefore, the total ripeness index of the berry can be expressed in the following way

(tab.113).

Ripening Index = 5,583-F1*0,528-F2*0,565+F3*0,136-F4*0,063


be concluded that the model used is of significant importance, that the Ripening Index

variable, is more relevant with the first and second constant rather than with the third and

fourth one (tab. 113).

B

Standard

deviation


F1 -,528 ,221 ,022

F2 -,565 ,159 ,001

F3 ,136 ,215 ,530

F4 -,063 ,171 ,714

Table 113. Coefficient values for the estimation of the total ripeness estimated of the berry,


Calculating the relation between the index ripeness expressed and that estimated statistically it

can be seen graphically how the ripeness expressed by the panel and that determined

statistically are overlapping (fig. 66).


152




In most cases there are values of Pearson's correlation positive, however the parameters

connected with the technological maturity of the grapes are negatively correlated.

The pH and anthocyanins‟s level per berry, are the parameters that don‟t present significant

Pearson‟s correlations (tab. 114).

153

Variable Bun.

wght

Ber.

wght °Brix

Titr.

Ac.

Rip

Ind.

Ber.

skin

wght

Ski.

%

Ber.

seed

num.

Seed

wght

/ber.

See.

%

Skin

anth.

Skin

poly.

Sees.

poly.

Tot.

polyp.

%

Skin

polyp.

Polyph

/berry F1 F2 F3 F4

Bunch

weight

0,60 0,88

Berry

weight

0,65 0,84 0,57 0,61 0,63 0,77 0,76 0,70 0,67

°Brix 0,65 0,60 0,58

Titratable

acidity

0,86

Ripening

index

0,86 0,63

Berry skin

weight

0,84 0,60 0,59 0,70 0,61 0,62 0,71

Skins % 0,63

Berry seed

number

0,57 0,59 0,61 0,66

Seed

weight/berry

0,61 0,70 0,61 0,63 0,57 0,87 0,68

Seeds % 0,06 0,63 0,91

Skin

anthocyanins

0,63 0,63 0,60

Skin

polyphenols

0,77 0,61 0,85

Seed

polyphenols

0,86 0,60 0,82

Total

polyphenols 0,60 0,76 0,85 0,86 0,59

% Skin

polyphenols

0,60 0,87

Polyphenols/

berry

0,70 0,62 0,57 0,64


154

Discriminant analysis on the technological and phenolic characteristics of the grapes at harvest‟s time

highlighted differences among clones examined and some of which may be distinct (fig. 67).


115).


variance % cumulated

Canonical correlation

1 21,472 68,5 68,5 0,977 2 6,831 21,8 90,3 0,934 3 2,121 6,8 97 0,824 4 ,937 3 100 0,696



points relative to the groups show a limited dispersion. Three distinct groups appear in the

centroids: the first comprises the thesis BM2, BM3 e BM4 that intersect each others. The

second BM1 and the third BM5 which well differs from the other, because well detached from the

other theses.

The 100,0% of the original grouping are classified correctly, while 84,4% of the cases


155


time.

Thesis

Group expected

Totals 1 2 3 4 5 Cross-

validation a % 1 100,0 ,0 ,0 ,0 ,0 100,0

2 ,0 77,8 11,1 11,1 ,0 100,0

3 ,0 11,1 88,9 ,0 ,0 100,0

4 ,0 11,1 11,1 77,8 ,0 100,0

5 11,1 ,0 11,1 ,0 77,8 100,0


a. Cross validation is done only for those cases in the analysis In cross validation, each case is



156

3.9 Aroma characteristics of the grapes at harvest’s time

The analysis performed by GC-MS allowed the identification of 220 aromatic compounds;

subdivided into 147 generated by enzymatic hydrolysis and 73 by acid hydrolysis.

Among these, only compounds present in significant quantities underwent statistical analysis

and they were divided into their belonging classes (tab. 117-118).

Ratios between compounds which are reported in literature as varietal ratios of the

„Sangiovese‟ cultivar were also studied (tab. 119).

Acids Aldehydes Benzene derivates

isocrotonic acid nonanal benzaldehyde

hexanoic acid citral methyl benzoate

hexanoic acid, 2-ethyl Aliphatic alcohols acetophenone

2-hexenoic acid isoamyl alcohol ethyl benzoate

sorbic acid 1-pentanol methyl salicylate

nonanoic acid 2-buten-1-ol, 3-methyl benzaldehyde, 2,5-dimethyl

n-decanoic acid 1-hexanol 1-phenylethanol

myristic acid 3-hexen-1-ol benzyl alcohol

pentadecanoic acid trans-2-Hexenol 2-phenylethanol

2,6 dimetil 6-hydroxy-2,7

octadienoic-acid

1-octen-3-ol

benzenepropanol

hexadecanoic acid octanol β-phenoxyethyl alcohol

stearic acid 4-octen-2,7-diol 2-(4-methoxyphenyl)ethanol

oleic acid 6-methoxy-3-methylbenzofuran

linoleic acid benzoic acid

abscisic acid 3',5'-dimethoxyacetophenone

3,4-dimethoxybenzyl alcohol

cinnamic acid

2,3,4-trimethoxybenzyl alcohol

Table117a.Compounds released by enzymatic hydrolysis (heterosides, ET1).

157

Esters Norisoprenoids

dimethyl succinate actinidol A

palmitic acid, methyl ester actinidol B

methyl palmitoleate 3,4-diidro-3-oxo- -ionol I

methyl stearate 3,4-diidro-3-oxo- -ionol II

methyl linoleate 3,4-diidro-3-oxo- -ionol III

methyl linolenate 3-hydroxy- β-damascone

methyl n-pentadecanoate 3-oxo- -ionol

fumaric acid, ethyl 2-methyl allyl ester

2,3-dehydro-4-oxo-7,8-dihydro- β-ionone

Monoterpenols methyl- β-ionone

linalool oxide A blumenol C

linalool oxide B 3-hydroxy-7,8-dihydro- β-ionol

linalool vomifoliol

-terpineol 7,8 dihydrovomifoliol

linalool oxide C (epoxylinalol) Phenols

linalool oxide D (epoxylinalol) guaiacol

citronellol phenol

myrtenol 4-vinylguaiacol

nerol eugenol

isogeraniol methoxyeugenol

geraniol phenol, 3,4,5-trimethoxy

exo-2-hydroxycineole coniferol 1

p-mentha-1,8-dien-6-ol coniferol 2

diol 1 Vanillins

p-cymen-7-ol vanillin

diol 2 methyl vanillate

2,3-pinanediol acetovanillone

trans-8-hydroxy-linalool homosyringic acid

p-menth-8-en-3-ol (isopulegol) zingerone

cis-8- hydroxy-linalool homovanillic alcohol

geranic acid 3,4,5-trimethoxybenzyl alcohol

p-menth-1-ene-7,8-diolo homovanillic acid

acetosyringone

Table117b.Compounds released by enzymatic hydrolysis (heterosides 1, ET1).

158

Alcohols monoterpenols Norisoprenoids

linalool trimethyl-dihydro-naphtalene (TDN)

1-terpinenol calamenene

4-terpineol -calacorene

hotrienol TDN 2

myrcenol 1,1,6,8-tetramethyl-1,2-dihydro-naphthal

cis-β-terpineol naphthalene, 1,4,6-trimethyl-

ocimenol 1 biphenyl, 4-isopropyl-

-terpineol cadalene

-terpineol Alcohols + ethers norisoprenoids

ocimenol 2 vitispiran 1

2-cyclohexene-1-methanol, 2,6,6-trimethyl- vitispiran 2

citronellol riesling acetale

nerol lanceol, cis

geraniol actinidol A

exo-2-hydroxycineole actinidol B

p-menth-1-en-9-ol OH-TDN

p-mentha-1,8-dien-6-ol -eudesmol

p-mentha-1,4-dien-7-ol -cadinol

Hydrocarbon monoterpenols -ionol

-terpin 1,2-naphthalenediol, 1,2,3,4-tetrahydro

-terpin 1-naphthalenol, 1,2,3,4-tetrahydro

terpinolene Ketones norisoprenoids

spiro[4.4]nona-1,6-diene 4-(2,4,4-trimethyl-cyclohexa-1,5-dienyl)

Oxides monoterpenols β-damascenone

2H-pyran, 2-ethenyltetrahydro-2,6,6-

trimethyl

ethanone, 1-(2,3-dihydro-1,1-dimethyl-1H

1,4-cineol

4-(2,6,6-trimethylcyclohexa-1,3-dienyl)but-

3-en-2-one

furan, tetrahydro-2,2-dimethyl mansonone C

trans-rose oxide 1.4,4,5,8-tetramethyl-4H-chromene

cis -rose oxide 4-(2,3,6-trimethylphenyl)-2-butanone

limonene oxide megastigmatrienone

linalool oxide A

1,4-hexadien-3-one, 5-methyl-1-[2,6,6-

trimethyl-2,4-cyclohexadien-1-yl]-

linalool oxide B 3,3,5,6-tetramethyl-1-indanone

2H-pyran, 3,6-dihydro-4-methyl-2-(2-

methyl-1-propenyl)-(nerol oxide 1)

2H-pyran, 3,6-dihydro-4-methyl-2-(2-

methyl-1-propenyl)-(nerol oxide 2)

3,6-dimethyl-2,3,3a,4,5,7a-

hexahydrobenzofuran

Table 118.Compounds released by acid hydrolysis (heterosides 2, ET2).

159

Linalool oxA/linalool oxB >1 (enzymatic hydrolysis)

Linalool oxC/linalool oxD >1 (enzymatic hydrolysis)

Linalool/geraniol < 1 (enzymatic hydrolysis)

Trans 8-OH Linalool/cis 8-OH linalool >1 (enzymatic hydrolysis)

Trans 8-OH Linalool + cis 8-OH linalool/p-menth-1en-7,8-diol >1

(enzymatic hydrolysis)

3-hydroxy β-damascenone/3-oxyde α-ionol <1 (enzymatic hydrolysis)

Linalool oxA/linalool oxB >1 (acid hydrolysis)

Table 119. Ratios between compounds.

The aromatic compounds, underwent MANOVA, factorial, discriminating and correlating

statistical analysis (tab. 120).

Abbreviation Variable M.U.

Aliph. Alc. ET1 Aliphatic alcohols (enzymatic hydrolysis) ng/g

Der. Benzene ET1 Benzene derivates (enzymatic hydrolysis) “

Phenols ET1 Phenols (enzymatic hydrolysis) “

Vanillins ET1 Vanillins (enzymatic hydrolysis) “

Monoterp. ET 1 Monoterpenols (enzymatic hydrolysis) “

Norisopren. ET1 Norisoprenoids (enzymatic hydrolysis) “

Aldehydes ET1 Aldehydes (enzymatic hydrolysis) “

Acids ET1 Acids (enzymatic hydrolysis) “

Esters ET1 Esters (enzymatic hydrolysis) “

Hydro. monot. ET2 Hydrocarbon monoterpenols (acid hydrolysis) “

Oxi. monot. ET2 Oxides monoterpenols (acid hydrolysis) “

Alc. monot. ET2 Alcohols monoterpenols (acid hydrolysis) “

Hydr. norisopr. ET2 Hydrocarbon norisoprenoids (acid hydrolysis) “

Ket. norisopr. ET2 Ketones norisoprenoids (acid hydrolysis) “

Alc. norisopr. + Ether ET2 Alcohols + ethers norisporenoids (acid hydrolysis) “

Heterosides 1 Compounds released by enzymatic hydrolysis “

Heterosides 2 Compounds released by acid hydrolysis “

Total Compounds released by enzymatic and acid hydrolysis “

V158 Linalool oxA /linalool oxB (enzymatic hydrolysis) n

V159 Linalool oxC /linalool oxD “

V161 Linalool/geraniol “

V162 Trans 8-OH linalool/cis 8-OH linalool “

V165 Trans 8-OH linalool + cis 8-OH linalool/p-menth -1ene-7,8-

diol “

V166 3-hydroxy β-Damascenone/3-oxide α-ionol “

V167 Linalool oxA /linalool oxB (acid hydrolysis) “

Table 120. List of abbreviations used in the text.

160

The MANOVA analysis variance was conducted studying the significance of the variables in

function of the factor choice (tab. 121).

Factor Dipendent variable F Sign.

Year Aliph. alc. ET1 0,371 ,691

Der. benzene ET1 29,313 ,000

Phenols ET1 2,738 ,069

Vanillins ET1 5,599 ,005

Monoterp. ET 1 2,413 ,095

Norisopren. ET1 11,626 ,000

Aldehydes ET1 8,097 ,001

Acids ET1 138,683 ,000

Esters ET1 9,283 ,000

Hydro. monot. ET2 23,965 ,000

Oxi. monot. ET2 27,172 ,000

Alc. monot. ET2 6,778 ,002

Hydr. norisopr. ET2 17,320 ,000

Ket. norisopr. ET2 15,929 ,000

Alc. norisopr. + Ether ET2 4,639 ,012

Heterosides 1 8,286 ,000

Heterosides 2 26,103 ,000

Total 26,103 ,000

V158 9,351 ,000

V159 8,722 ,000

V161 35,769 ,000

V162 43,580 ,000

V165 1,498 ,228

V166 9,688 ,000

V167 43,963 ,000


In the interaction thesis by year, all the aromatic classes of the compounds (tab. 120) and the

ratios between aromatic compounds, resulted statistically notable, therefore they have not

been recorded on the table (tab. 121). However, if the year is the factor in the statistic

analysis, some parameters are not statistically important, i.e: aliphatic alcohols, phenols,

monoterpenols from enzymatic hydrolysis and the ratio trans 8-OHlinalool+cis8-

OHlinalool/p-menth-1 en-7,8-diol.

Using data previously obtained the amount of the variability attributed to the different factor

was calculated (tab. 122).

161

Variability is equally attributable to the thesis and to the year as shown in the table 6, more

precisely, the compounds, originating from enzymatic hydrolysis except for aldehydes and

acids, and 3- hydroxy β-damascenone/3-oxide α-ionol and linalool oxA/linalool oxB (acid

hydrolysis) have a greater variability due to the thesis, while the compounds originating from

acid hydrolysis and the other ratios are strongly influenced by the year effect. Aldehydes,

esters and the total of heterosides 1 and hydrocarbon monoterpenols and norisoprenoids for

the heterosides 2, show comparable levels of variability due to thesis and year (tab. 122).

Variable

%

variability

due to

the thesis

%

variability

due to

the year

%

variability

due to the

interaction

thesis/ year

%

variability

due to

error

Aliph. alc. ET1 74,53 1,75 19,01 4,72

Der. benzene ET1 51,86 37,56 9,29 1,28

Phenols ET1 44,97 11,84 38,87 4,32

Vanillins ET1 66,23 18,95 11,43 3,38

Monoterp. ET 1 52,83 20,05 18,81 8,31

Norisopren. ET1 31,20 55,47 8,56 4,77

Aldehydes ET1 40,39 37,95 16,98 4,69

Acids ET1 37,18 52,82 9,62 0,38

Esters ET1 41,30 42,03 12,14 4,53

Hydro. monot. ET2 47,77 44,25 6,13 1,85

Oxi. monot. ET2 34,90 57,56 5,42 2,12

Alc. monot. ET2 59,92 21,37 15,56 3,15

Hydr. norisopr. ET2 42,38 47,12 7,77 2,72

Ket. norisopr. ET2 27,62 56,29 12,55 3,53

Alc. norisopr. + Ether ET2 64,49 20,91 10,09 4,51

Heterosides 1 37,79 37,51 20,17 4,53

Heterosides 2 27,05 62,65 7,90 2,40

Total 27,05 62,65 7,90 2,40

V158 38,00 39,01 18,82 4,17

V159 22,75 42,49 29,89 4,87

V161 33,03 56,29 9,11 1,57

V162 17,53 70,43 10,42 1,62

V165 37,38 19,93 29,38 13,31

V166 47,28 15,62 35,49 1,61

V167 11,25 80,28 6,64 1,83


162

Using Tukey test, it is possible to note that all the parameters analyzed have given rise to

homogenous differentiated subsets (tab. 123-125).

The significant variables that create less differentiated homogenous subsets are the

compounds from the esters classis obtained from the enzymatic way and the ratio linalool

oxA /linalool oxB obtained, instead, from the acid hydrolysis.

Within the compounds freed by the enzymatic way the families of the benzene derivatives and

of the norisoprenoids classes are more present in theses while the aldehydes are quantitatively

lower; from a concentration of hundreds of ng/g fresh vegetal tissue weight to values close to

the unit (tab. 123).

Hydrocarbon norisoprenoids and monoterpenols are the classes that are more and less

represented in the hydrolyzing of aromatic compounds by acid way (tab. 124).

Comparing the theses, the „Chianti Colline Pisane‟ thesis is quite distinct from the others for

the quantitative inferior values of most of the aromatic classes. Samples from the province of

Siena, however, have a higher aroma content: benzene derivatives, phenols, vanillins, and

aldehydes. As for enzymatic hydrolysis and hydrocarbon and monoterpenol oxides for the

acid one, are more present in the „Chianti Classico‟ area while the rest are found in the area of

the „Brunello di Montalcino‟. In the province of „Grosseto‟,the „Morellino di Scansano‟ thesis

differs from the others for the highest levels of monoterpenol alcohols, while two thesis from

„Montecucco‟ for the lowest in phenols and aldehydes. The benzene derivatives class is that

with the most variability within the theses. Analyzing the total quantity of aromatic

compounds, those freed by enzymatic hydrolysis are more present in a thesis of the „Brunello

di Montalcino‟ and scarcely present in a „Montecucco‟ thesis, while those freed by acid

hydrolysis are more present in the province of Siena and more precisely in the „Chianti

Classico‟ and in a limited way in the „Chianti Colline Pisane‟ thesis (tab. 123-125).

As can be seen from all the ratios examined there is perfect harmony in the results and in the

values obtained by other authors for the „Sangiovese‟ cultivar (Di Stefano, 1998 et Lanati

2001) The first and second ratio regarding the linalool oxides are completely favourable to the

trans form compounds compared to the cis one; the linalool/geraniol ratio is much less than

one in the grapes with a low linalool concentration (tab. 125).

163

Thesis Aliph.

alc .

Benz

der

Phenols Vanillin Monoterp Norisopr. Aldehyd Acids Esters

CCP 1 51,54 a 281,29 a 44,80a 74,36 a 62,98 a 138,22 a 1,57 ab 55,41abc 14,03a

MC 1 140,95 def 481,98 abc 64,00 a 126,85abc 114,62 a-d 247,61 abc 1,15 a 71,87 ad 11,79a

MC 2 121,62 cd 423,82ab 38,04 a 99,19 ab 68,54 a 168,92 a 2,49 bc 77,45 ad 12,54a

MC 3 173,35 f 606,54bcd 96,92ad 164,31 be 111,47 a-d 297,80 a-e 2,84 c 106,38cd 10,95a

MC 4 164,52 ef 586,38 bcd 92,39abc 166,82 be 111,35 a-d 274,33 a-d 2,28 abc 114,98de 23,26a

MC 5 148,38def 522,82 abc 68,05 a 140,78 ad 106,16 a-d 205,01 abc 1,34 ab 84,11 ad 31,92ab

MC 6 95,50bc 592,49 bcd 175,16de 186,32cde 135,44cd 359,33 b-e 6,14 d 48,8 ab 17,65 a

MS 1 133,26 cde 800,35 de 207,40 e 186,24cde 114,22 a-d 435,53de 6,34 d 39,52 a 20,59 a

BM 1 160,58def 441,89 ab 65,19 a 125,95abc 91,09 abc 242,10 abc 2,22 abc 101,77bcd 15,60 a

BM 2 140,38 def 437,87 ab 59,21 a 120,85abc 66,11a 165,83 a 2,89 c 52,98 abc 11,04 a

BM 3 151,84 def 471,26 abc 52,60 a 111,11abc 75,28 ab 220,42 abc 1,50 ab 86,85 a-d 11,85 a

BM 4 132,60 cde 504,47 abc 71,27 ab 117,05abc 72,52ab 234,71 abc 2,49 bc 79,96 a-d 8,45 a

BM 5 268,26 g 815,22 de 115,79ad 213,84 de 147,66 d 463,41 e 3,42 c 162,38 e 13,38 a

BM 6 96,15 bc 695,46 cde 206,09 e 185,23 be 123,46bcd 481,85 e 8,11 e 39,21 a 5,38a

BM 7 74,96 ab 501,81abc 157,06cde 230,41 e 136,10 cd 388,56 cde 7,03 de 115,72 de 73,12 b

CC 1 132,28 cde 888,43 e 288,65 f 183,45 be 96,65 a-d 377,42 b-e 8,14 e 60,62 abc 26,62 a

CC 2 68,50 ab 575,22bcd 150,32 be 187,16cde 71,27ab 201,86 ab 6,42 d 68,84 a-d 17,56 a

Table 123. Significant parameters with different subsets; heterosides 1. Tukey Test (p=0,05).

Thesis

Hydroc.

monot.

Oxid.

monot.

Alc.

monot.

Hydr.

norisopr.

Ket.

norisopr.

Alc.

norisopr. +

Ethers

CCP 1 0,21 ab 2,19 ab 2,02 a 6,65a 2,42 a 10,61 a

MC 1 0,46 b-g 4,12 de 4,76 b-g 31,01d-g 4,66 ab 30,89 de

MC 2 0,41 a-e 4,08 cde 3,72 ab 26,91 b-g 7,78 a-d 21,78 a-d

MC 3 0,27 abc 3,30 bcd 4,65 b-f 13,95a-d 5,60abc 15,58ab

MC 4 0,26 ab 3,17 a-d 4,07 bcd 19,95 a-f 5,39 abc 16,98 abc

MC 5 0,21 ab 2,31 abc 3,33 ab 11,64 ab 2,55 a 12,06 a

MC 6 0,73 gh 6,06 f 6,43 fg 32,60 efg 9,70 bcd 33,59 e

MS 1 0,58 c-g 3,70 bcd 6,58 g 17,42 a-e 11,90 de 19,57 a-d

BM 1 0,36 a-d 2,86 a-d 3,19 ab 21,10 a-g 6,29 a-d 19,58 a-d

BM 2 0,12 a 3,24 bcd 3,03 ab 22,28 a-g 6,21 a-d 15,67 ab

BM 3 0,42 a-f 3,06 a-d 3,03 ab 19,79 a-f 5,10 abc 16,20 ab

BM 4 0,24 ab 1,40 a 1,97 a 12,66 abc 3,63 a 10,20 a

BM 5 0,72 fgh 4,13 de 5,59 c-g 36,15 fg 10,64 cd 36,62 e

BM 6 0,69 fgh 3,75 bcd 4,31 b-e 22,20 a-g 5,75 abc 29,92 de

BM 7 0,62 d-g 3,72 bcd 3,87 bc 30,29 c-g 6,73 a-d 25,68 b-e

CC 1 1,01 h 5,78 ef 5,91 d-g 39,61 g 17,23 e 29,82 de

CC 2 0,75gh 7,78 g 5,96 efg 39,74 g 16,67 e 28,01 cde


164

Thesis

Heter.

1

Heter.

2

Total V158 V159 V161 V162 V165 V166 V167

CCP 1 1155,23ab 24,22 a 1179,45abc 1,15 a 8,22 bc 0,14 f 1,43 a 11,75de 0,05abc 2,55 a

MC 1 1299,30abc 75,9 dg 1375,23 ad 1,29ab 6,95abc 0,04ab 2,68 e 2,25 ab 0,03 ab 2,49 a

MC 2 1046,24 a 64,70bg 1110,94 a 1,26ab 9,51 c 0,07ad 2,76 e 1,76 a 0,06 ad 3,19 a

MC 3 1624,16 af 43,37ad 1667,54 af 1,32abc 7,63abc 0,0 abc 2,48 de 1,68 a 0,05 ad 2,86 a

MC 4 1583,01 ae 49,84ad 1632,85 ag 1,40 ad 8,64 bc 0,05 bc 2,22 be 1,87ab 0,06 ad 2,91a

MC 5 1349,30abc 32,12abc 1381,42 ad 1,41 ad 7,48abc 0,04 ab 2,65 e 1,86 ab 0,04abc 2,73 a

MC 6 1688,44 af 89,1 efg 1777,58 bg 1,33abc 6,31 ab 0,12def 1,55 ab 5,50 bc 0,03 ab 2,94 a

MS 1 2010,67def 59,76 bf 2070,43efg 1,41 ad 6,40 ab 0,12 ef 1,36 a 2,51 ab 0,03 ab 2,59a

BM 1 1304,84abc 54,54 ae 1359,38 ad 1,42 ad 6,77abc 0,08 be 1,63abc 1,95 ab 0,05 ad 4,67 a

BM 2 1098,24ab 50,59 ad 1148,86 ab 1,59 cd 7,89abc 0,05abc 2,23 be 1,93ab 0,09 d 2,77 a

BM 3 1225,09 ab 47,69 ad 1272,77abc 1,47bcd 8,70 bc 0,04 ab 2,18 be 1,40 a 0,06 ad 4,97a

BM 4 1260,88 ab 30,12ab 1290,99abc 1,45bcd 8,44bc 0,05 ab 1,74 ad 1,50 a 0,07 cd 2,79 a

BM 5 2277,13 f 93,86 fg 2370,99 g 1,32abc 8,07abc 0,03 a 2,2 cde 2,29 ab 0,04abc 2,62a

BM 6 1927,83 cf 66,63 cg 1994,46 dg 1,47bcd 7,28abc 0,05abc 1,87 ad 2,20 ab 0,02 a 39,46b

BM 7 1754,55 bf 70,93 dg 1825,48 cg 1,68 d 5,32 a 0,06abc 2,46 de 2,08ab 0,05abc 36,03b

CC 1 2157,71 ef 99,38 g 2257,09 fg 1,47bcd 8,61 bc 0,08 a-e 1,22 a 14,77e 0,02 a 70,54b

CC 2 1394,84 ad 98,93 g 1493,78 ae 1,51bcd 12,53d 0,01 cf 1,31 a 8,70 cd 0,06bcd 2,75a

Table 125. Significant parameters with different subsets; heterosides 1, 2 and ratios. Tukey Test

(p=0,05).

From the multivariate analyses, choosing the year as the factor and the Tukey test, it emerged

that most of the parameters examined originated differentiated subsets in particular in the year

2011 (tab. 126). The year 2010 that stands out as having the highest level of aromatic families

with the exception of aliphatic alcohols (enzymatic hydrolysis) which were quantitatively

higher the year before. The aromas found in the 2011 harvest were mainly the phenols,

vanillins, aldehydes, esters, hydrocarbon and ketones norisoprenoids classes even if the values

were quantitatively similar to those of the other years.

Variable 2009 2010 2011

Aliphatic alcohols ET1 145,40 b 140,13 b 112,36 a

Benzene derivates ET1 549,54 a 605,50 a 543,90 a

Phenols ET1 99,23 a 114,12 ab 131,28 b

Vanillin ET1 141,28 a 158,10 a 162,06 a

Monoterpenols ET 1 94,53 a 117,60 b 88,74 a

Norisoprenoids ET1 251,96 a 337,91 b 275,35a

Aldehydes ET1 2,80 a 3,66 b 5,26 c

Acids ET1 73,10 a 96,20 b 71,92 a

Esters ET1 13,70 a 17,24 a 26,54 a

Hydroc. monot. ET2 0,32 a 0,56 b 0,55 b

165

Variable 2009 2010 2011

Oxid. monot. ET2 3,61 a 4,25 b 3,55 a

Alc. monot. ET2 3,78 a 4,10 b 4,01 a

Hydr. norisopr. ET2 16,96 a 26,68 b 27,80 b

Ket. norisopr. ET2 6,56 a 7,42 ab 8,65 b

Norisopr. alc.+ Ether ET2 13,05 a 27,12 b 25,61 b

Heterosides 1 1444,55 a 1722,22 b 1449,24 a

Heterosides 2 44,32 a 71,04 b 70,24 b

Total 1488,88 a 1793,27 b 1519,48 a

V158 1,34 a 1,49 b 1,40 a

V159 1,34 a 1,49 b 1,40 a

V161 0,08 b 0,09 b 0,04 a

V162 1,87 b 1,62 a 2,53 c

V165 5,00 b 3,33 a 3,32 a

V166 0,06 b 0,03 a 0,06 b

V167 7,56 a 10,93 ab 14,84 b


horizontal.

Using the factorial statistics analysis it was possible to collect all the variables found and

calculated into four new complex variables (components) to represent 96,62% of the total

variability of the aromatic characteristics of the berries at harvest (tab. 127). The first

component is linked to most of the variables studied: to the variables that indicate varietal

ratios except for linalool/geraniol, to many of the families belonging to the heterosides 1, to

the total, between the heterosides 2, to the hydrocarbon and alcohols monoterpenols. The sum

of linalool/geraniol, the aromatic classes freed by acid hydrolysis not previously mentioned

and the aliphatic alcohols from the enzymatic hydrolysis characterize the second component.

The third is linked to the esters and to the acids from the heterosides 1, and the fourth, only to

vanillin from the enzymatic hydrolysis (tab. 128).

Component


Total

%

variability

%

cumulated

1 15,39 51,16 51,16

2 9,38 31,26 82,42

3 2,35 7,84 90,26

4 1,91 6,36 96,62

Table 127. Results of the factorial analysis: pricipal components method.

166

Component

1 2 3 4

Total ,987

Ketones norisoprenoids ET2 ,983

Hydrocarbon norisoprenoids ET2 ,980

Heterosides 2 ,979

Aliphatic alcohols ET1 -,863

Esters ET1 ,939

Heterosides 1 ,986

Benzene derivates ET1 ,972

Norisoprenoids ET1 ,935

V162 ,935

V167 ,902

Phenols ET1 ,882

Aldehydes ET1 ,873 -,402

Monoterpenols ET 1 ,741 ,464

Alcohols monoterpenols ET2 ,713 ,499

Hydrocarbon monoterpenols ET2 ,702 ,676

Vanillin ET1 ,479 ,511 ,687

Acids ET1 ,465 ,440 ,702

Alcohols + ethers norisoprenoids ET2 ,451 ,804

V161 -,490 ,858

Oxides monoterpenols ET2 -,570 ,755

V165 -,789 -,461

V159 -,856 ,404

V158 -,875

V166 -,986



By extracting only two of the components from the total data it is possible to explain 87,49%

of the total variability (tab. 129). The first component is linked to the variables that represent

the ratios between specific aromas and those of the heterosides 1, while the second to the

heterosides 2 (tab. 130).

On the graph (fig. 68) it is possible to see how the descriptors that are found in the same

quadrant are directly correlated while those further away are correlated negatively. Both

components are negatively correlated to the ratio trans 8-OH linalool+cis 8-OH linalool/p-

menth-1ene-7,8-diol. Furthermore, the first component is negatively correlated to other ratios

between specific components except trans 8-OH linalool/cis 8-OH linalool and linalool oxA/

linalool oxB coming from acid hydrolysis, the esters of the heterosides 1, the total of the

components obtained by acid hydrolysis and the hydrocarbon norisoprenoids and oxides

monoterpenols classes. The second component, is negatively correlated with the sum of the

167

freed components by enzymatic hydrolysis, aliphatic alcohols, benzene derivatives, aldehydes

and the trans 8-OH linalool/cis 8-OHlinalool (fig. 68).

Component


Total

%

variance

%

cumulated

1 15,51 51,69 51,69

2 10,74 35,81 87,49



Variable Component

1 2

Total ,998


Heterosides 2 ,949



Esters ET1 ,555

Heterosides 1 ,997



V162 ,923

V167 ,913

Phenols ET1 ,906

Aldehydes ET1 ,850 -,457

Monoterpenols ET 1 ,798 ,441



Vanillin ET1 ,545 ,672

Acids ET1 ,525 ,670

Alcohols + ethers norisoprenoids

ET2 ,470 ,858

V161 -,492 ,830


V165 -,801 -,493

V159 -,818

V158 -,853 ,460

V166 -,978



168


Considering the correlation between the parameters analyzed, only the variables characterized

by a probability < 0,00001 were reported on the table. Pearson‟s significant correlations are

mostly positive while those linked to the variables expressing the ratios between single

aromatic components are negative. Correlations close to the unit value are present between

the value of the heterosides 1 and the content of the benzene derivatives and norisoprenoids

while the heterosides 2 are strictly linked to those of hydrocarbon norisoprenoids and alcohols

plus ether monoterpenols classes. Pearson‟s correlations among ratios and others variables

aren‟t so significant (tab. 131).

169

Aliph.

Alc

ET1

Benz.

Der.

ET1

Phen

ET1

Van.

ET1

Mon.

ET 1

Nor.

ET1

Ald.

ET1

Acid

ET1

Est.

ET1

Hydr.

Monot

ET2

Oxid

Monot.

ET2

Alc.

Monot

ET2

Hydr.

Noris.

ET2

Ket.

Noris.

ET2

Alc+Et.

Noris.

ET2

Heter.

1

Heter.

2 Tot.

Aliph. Alc.ET1 0,37 0,54 0,34 0,33

Benz. Der

ET1

0,37 0,78 0,62 0,42 0,78 0,39 0,31 0,92 0,92

Phenols

ET1

0,78 0,73 0,43 0,79 0,64 0,52 0,39 0,45 0,31 0,83 0,32 0,83

Vanillins

ET1

0,62 0,73 0,58 0,77 0,40 0,32 0,39 0,32 0,77 0,78

Monoterpenols

ET 1

0,42 0,43 0,58 0,58 0,40 0,32 0,40 0,38 0,59 0,60

Norisoprenoids

ET1

0,78 0,79 0,77 0,58 0,47 0,40 0,34 0,90 0,90

Aldehydes

ET1

0,39 0,64 0,40 0,47 0,49 0,35 0,44 0,38 0,41 0,37 0,43

Acids ET1 0,54 0,32 0,40 0,33 0,37 0,36

Esters ET1 0,33

Oxid.

Monot. ET2

0,35 0,61 0,71 0,72 0,66 0,59 0,76

Alcohols

Monot. ET2

0,39 0,40 0,65 0,71 0,61 0,63 0,67 0,73

Hydrocarbon

Norisop.ET2

0,69 0,72 0,61 0,73 0,84 0,97

Ketones

Norisop.ET2

0,45 0,44 0,61 0,66 0,63 0,73 0,55 0,77 0,26

Alcoh.+ Eters

Norisop.ET2

0,31 0,32 0,38 0,34 0,38 0,77 0,59 0,67 0,84 0,55 0,92 0,32

Heterosides 1 0,34 0,92 0,83 0,77 0,59 0,90 0,41 0,37 0,38

Heterosides 2 0,32 0,37 0,78 0,76 0,73 0,97 0,77 0,92

Total 0,33 0,92 0,83 0,78 0,60 0,90 0,43 0,36 0,42 0,34 0,32

Table. 131 Pearson‟s correlations. Legend abbreviations is on the previous pages.

170


analysis of the most relevant variables for statistics purposes were inserted in stepwise (tab.

132).

Aliphatic alcohols (enzymatic hydrolysis)

Linalool oxA/linalool oxB (acid hydrolysis)

Trans 8-OH linalool + cis 8-OH linalool/p-menth -1ene-7,8-diol

Ketones norisoprenoids (acid hydrolysis)

Compounds released by enzymatic hydrolysis

Benzene derivates (enzymatic hydrolysis)

Aldehydes (enzymatic hydrolysis)

Vanillins (enzymatic hydrolysis)

Table 132. Variables inserted in the analysis.

Discriminating analysis on the aroma characteristics of the berries at harvest has highlighted

differences in the winegrowing areas examined, some of these are distinct (fig. 64). In

particular the first two functions explain over 60% of total variability (tab. 133).

Function Eingenvalue

% of

variance

%

cumulated

Canonical

correlation

1 34,702 48,6 48,6 ,986

2 10,145 14,2 62,8 ,954

3 7,885 11,0 73,9 ,942

4 5,592 7,8 81,7 ,921

5 3,680 5,2 86,9 ,887

6 2,809 3,9 90,8 ,859

7 2,193 3,1 93,9 ,829

8 1,541 2,2 96,0 ,779

9 1,231 1,7 97,8 ,743

10 ,773 1,1 98,8 ,660

11 ,363 ,5 99,3 ,516

12 ,170 ,2 99,6 ,381

13 ,155 ,2 99,8 ,367

14 ,076 ,1 99,9 ,265

15 ,065 ,1 100,0 ,247


171

Looking at the graph (fig. 69) of the centroids obtained from the discriminating analysis data

subdivided by area (tab. 134), one thesis in the „Chianti Classico‟ area stands out. The other

thesis are close to one another and some points overlap. The two „Montecucco‟ theses do not

appear very close differently from those of the „Brunello and Montalcino‟. In the „Chianti

Classico‟ the two theses are very different. 100% of the original cases grouped are correctly

classified (tab. 135).

Area Denomination area Estate

1 Colline Pisane Beconcini

2 Montecucco Collemassari

3 Montecucco Salustri

4 Morellino di Scansano Fattoria di Magliano

5 Brunello di Montalcino Col D'Orcia

6 Brunello di Montalcino Casanova di Neri

7 Brunello di Montalcino La Mannella

8 Chianti Classico Capannelle

9 Chianti Classico Castello di Albola

Table 134. Area subdivision.

Figure 69. Centroids obtained from the cluster analysis of the aromatic characteristics of the

grapes at harvest‟s time.

Area

172

Thesis

Group expected

Total 1 2 3 4 5 6 7 8 9

Cro

ss-v

alid

atoa

% 1 100,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0

2 ,0 100,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0

3 ,0 ,0 100,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0

4 ,0 ,0 ,0 100,0 ,0 ,0 ,0 ,0 ,0 100,0

5 ,0 ,0 ,0 ,0 100,0 ,0 ,0 ,0 ,0 100,0

6 ,0 ,0 ,0 ,0 ,0 100,0 ,0 ,0 ,0 100,0

7 ,0 ,0 ,0 ,0 ,0 ,0 100,0 ,0 ,0 100,0

8 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0 ,0 100,0

9 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,0 100,0 100,0




b. 100,0% of original grouped cases correctly classified.


Statistics analysis was also carried out on the aromatic compounds to study the characteristics

and possible important statistical differences of the grapes from the theses belonging to the

same Denomination of Origin. This is the case of to „Montecucco‟ area where multivariate,

factorial, discriminating and correlation analyses were carried out.

In addition, among the theses coming from the area of „Montalcino‟ was also studied the case

of „Col D'Orcia‟ estate in order to study in detail the clone effect grown in the same site of

cultivation.

173

3.9.1. ‘Montecucco’ area

MANOVA analysis was conducted studying the significance of the variables in function of

the factor choice (tab. 136).

Factor Dipendent variable F Sig. Factor Dipendent variable F Sig. Year Aliph. alc. ET1 0,22 ,804 Thesis*

Year Aliph. alc. ET1 1,07 ,413

Der. benzene ET1 52,55 ,000 Der. benzene ET1 5,69 ,000

Phenols ET1 7,81 ,002 Phenols ET1 3,73 ,002

Vanillins ET1 0,00 ,998 Vanillins ET1 2,13 ,049

Monoterp. ET 1 0,68 ,514 Monoterp. ET 1 4,41 ,000

Norisopren. ET1 8,14 ,001 Norisopren. ET1 4,96 ,000

Aldehydes ET1 5,48 ,009 Aldehydes ET1 3,47 ,003

Acids ET1 101,53 ,000 Acids ET1 9,87 ,000

Esters ET1 0,16 ,850 Esters ET1 2,16 ,045

Hydro. monot. ET2 2,81 ,074 Hydro. monot. ET2 4,35 ,001

Oxi. monot. ET2 3,07 ,059 Oxi. monot. ET2 3,49 ,003

Alc. monot. ET2 14,21 ,000 Alc. monot. ET2 12,06 ,000

Hydr. norisopr.

ET2 21,51 ,000

Hydr.

norisopr. ET2 2,91 ,009

Ket. norisopr. ET2 8,49 ,001 Ket. norisopr. ET2 19,22 ,000

Alc. norisopr. + Ether

ET2 6,71 ,003

Alc. norisopr. +

Ether ET2 10,90 ,000

Heterosides 1 3,390 ,045 Heterosides 1 3,210 ,005


Total 12,57 ,000 Total 16,39 ,000

V158 3,17 ,054 V158 3,10 ,006

V159 19,79 ,000 V159 4,31 ,001

V161 29,33 ,000 V161 71,11 ,000

V162 33,80 ,000 V162 35,09 ,000

V165 2,09 ,139 V165 1,52 ,174

V166 48,16 ,000 V166 60,15 ,000

V167 27,50 ,000 V167 3,59 ,002


From MANOVA analysis, the only variable that, apart from the choice of factor, remains

statistically non significant is the ratio trans 8-OH linalool + cis8-OH linalool/p-menth-1-ene-

7,8-diol. Moreover this ratio is the only parameter statistically significant selecting the thesis

as the factor and to this, aliphatic alcohols from enzymatic hydrolysis are added if the

interaction thesis by year is indicated. Vanillins, monoterpenols and esters obteined from

enzymatic hydrolysis and hydrocarbon and monoterpenols oxides from hydrolysis acid, are

174

the aromatic classes that appear statistically non important when the factor is the year. Ratios

between single compounds, linalool oxA/linalool oxB lose importance with the year change.

Using data previously obtained, the amount of the variability attributed to the different factor,

was calculated (tab. 137). For most of the parameters the variability is attributable to the

thesis; the year, however, shows more variability as concerns benzene derivates, phenols,

acids and hydrocarbon norisoprenoids. The parameters linked to the ratios show comparable

levels of variability attributable to the different source, except of linalool oxA/linalool oxB

(acid hydrolysis) and linalool/geraniol (tab. 137).

Variable

%

variability

due to

the thesis

%

variability

due to

the year

%

variability

due to

interaction

thesis/year

%

variability

due to

error

Aliph. alc. ET1 79,61 1,95 9,51 8,92 Benzene der. ET1 18,40 72,38 7,84 1,38 Phenols ET1 17,31 51,50 24,60 6,60 Vanillins ET1 81,77 0,01 12,39 5,83 Monoterp. ET 1 52,46 5,29 34,45 7,80 Norisopren. ET1 43,76 32,48 19,77 3,99 Aldehydes ET1 47,85 28,70 18,21 5,24 Acids ET1 35,23 58,51 5,69 0,58 Esters ET1 54,06 2,25 29,88 13,81

Hydro. monot. ET2 58,89 14,16 21,91 5,03

Oxi. monot. ET2 61,06 15,80 17,98 5,16

Alc. monot. ET2 44,30 29,03 24,63 2,04

Hydr. norisopr. ET2 30,11 59,13 8,00 2,75

Ket. norisopr. ET2 42,72 16,95 38,34 2,00

Alc. norisopr. + Ether ET2 45,19 19,76 32,10 2,95 Heterosides 1 40,67 26,47 25,06 7,81 Heterosides 2 45,59 22,83 29,77 1,82 Total 45,59 22,83 29,77 1,82 V158 41,48 25,53 24,94 8,05 V159 9,02 71,74 15,61 3,63 V161 27,33 21,01 50,94 0,72 V162 32,34 32,72 33,97 0,97 V165 24,97 34,01 24,73 16,29 V166 36,83 27,83 34,76 0,58 V167 10,71 76,52 10,00 2,78


175

Using Tukey test, it is possible to note that all the parameters analyzed have given rise to

homogenous differentiated subsets (tab. 138-140).

The compounds originated by enzymatic way, except for benzene derivates and esters classis

and the compounds obtained by acid hidrolysis, create differentiated homogenous subsets

(tab. 138-139).

The variables describing the ratios create both type of homogenous subsets (tab. 140).

Regarding heterosides 1, the most quantitatively present class is that of the benzene

derivatives while the aldehydes and esters the least. The thesis with the highest values in most

of the different families is the same as the one having lower aliphatic alcohols and acid levels

and it is the only thesis not from the „ColleMassari‟ wine farm (tab. 138).

Among the compounds extractedby acid hydrolysis the most present are those from the

hydrocarbon norisoprenoids class and in a minor way those from monoterpenols oxides and

ketones norisoprenoids (tab. 138).

In addition, the grapes from the above mentioned thesis also stand out for the highest aroma

levels from heterosides 2, while the lowest levels belong to another single thesis (tab. 139).

The ratios examined accord perfectly well with each other and with the values obtained by

other authors on the „Sangiovese‟ cultivar (Di Stefano et al, 1998; Lanati et al., 2001) (tab.

140).

Thesis Aliph. alcohols

Phenols Vanil. Monot. Norisop. Aldeh. Acids Esters Benzene der.

MC 1 140,95 bc

64,00 ab

126,85 ab

114,62 bc

247,61 ab

1,15 a

71,87 ab

11,80 a

481,98 a

MC 2 121,62

ab 38,04 a

99,19 a

68,54 a

168,92 a

2,49 c

77,45 ab

12,55 a

423,82 a

MC 3 173,35 d

96,92 b

164,31 bc

111,47 b

297,80 bc

2,84 c

106,38 b

10,95 a

606,54 a

MC 4 164,52

cd 92,39 b

166,82 bc

111,35 b

274,33 b

2,28 bc

114,98 b

23,26 a

586,38 a

MC 5 148,38

bcd 68,05 ab

140,78 abc

106,16 b

205,01 ab

1,34 ab

84,11 ab

31,92 a

522,82 a

MC 6 96,51 a

182,14 c

192,39 c

142,11 c

376,85 c

6,01 d

51,10 a

18,58 a

612,49 a

Table 138. Significant parameters with different and no differentiated subsets; heterosides 1. Tukey

Test (p=0,05).

176

Thesis Hydroc.

monot. Oxid. monot.

Alcoh.

monot. Hydr.

norisopr. Ket. norisopr

Alc.

norisopr.

+ Ethers MC 1 0,47 b 4,12 b 4,76 b 31,01 d 4,66 ab 30,89 c

MC 2 0,41 ab 4,09 b 3,72 b 26,91 cd 7,78 cd 21,78 b

MC 3 0,27 ab 3,30 ab 4,65 ab 13,95 ab 5,60 bc 15,58 ab

MC 4 0,26 ab 3,17 ab 4,07 ab 19,95 bc 5,40 b 16,99 ab

MC 5 0,22 a 2,31 a 3,33 a 11,64 a 2,55 a 12,06 a

MC 6 0,72 c 5,84 c 6,46 c 29,03 d 8,11 d 31,37 c


Thesis

Heter.

1

Heter.

2

Total V159 V161 V162 V166 V158 V167 V165

MC 1 1299,30 ab

75,92 de

1375,23 ab

6,95 ab

0,04 a

2,69 b

0,04 ab

1,29 a

2,50 a

2,25 a

MC 2 1046,24 a

64,70 cd

1110,94 a

9,51 c

0,07 b

2,76 b

0,06 b

1,26 a

3,19 a

1,76 a

MC 3 1624,16 b

43,37 ab

1667,54 b

7,63

abc 0,06

ab 2,48

b 0,05

b 1,32

a 2,86

a 1,68

a MC 4 1583,01

b 49,83

bc 1632,85 b

8,64 bc

0,05 ab

2,22 ab

0,06 b

1,40 a

2,90 a

1,87 a

MC 5 1349,30 ab

32,12 a

1381,42 ab

7,49 ab

0,04 a

2,65 b

0,04 ab

1,41 a

2,73 a

1,86 a

MC 6 1755,93 c

81,55 e

1837,48 b

6,41 a

0,13 c

1,53 a

0,03 a

1,34 a

2,97 a

5,21 b

Table 140. Significant parameters with different and no differentiated subsets; heterosides 1,2 and

ratios. Tukey Test (p=0,05).

From the multivariate analysis, choosing the year as the source and from the Tukey test, it

emerged that most of the parameters examined originate differentiated subsets (tab. 141).

The grapes harvested in 2009 and those of the following year stand out in equal measure, for

the highest levels of the aromatic families, with the exception of vanillins, esters and ketones

norisoprenoids classes which were predominant in 2011. The variables indicating the ratios

between single compounds do not predominate in any one year in particular.

Variable 2009 2010 2011

Aliphatic alcohols ET1 174,25 c 150,63 b 97,87 a

Benzene derivates ET1 633,55 c 537,15 ab 436,54 a

Phenols ET1 89,92 a 89,95 a 85,54 a

Vanillins ET1 150,20 a 140,86 a 151,86 a

Monoterpenols ET 1 116,51 b 116,71 b 91,06 a

Norisoprenoids ET1 241,83 a 307,48 b 227,65 a

177

Variable 2009 2010 2011

Aldehydes ET1 1,50 a 2,10 b 4,37 c

Acids ET1 88,05 a 82,52 a 84,22 a

Esters ET1 8,99 a 13,27 a 33,07 b

Hydroc. monot. ET2 0,33 a 0,46 b 0,36 ab

Oxid. monot. ET2 4,51 c 3,71 b 3,03 a

Alc. monot. ET2 4,66 b 5,42 b 3,23 a


Ket. norisopr. ET2 6,14 b 4,59 a 6,22 b

Norisopr. alc.+ Ether ET2 14,17 a 28,40 c 21,20 b

Heterosides 1 1560,67 b 1491,68 ab 1248,42 a

Heterosides 2 47,86 a 67,23 b 57,30 a

Total 1608,54 b 1558,91 ab 1305,73 a

V158 1,24 a 1,30 a 1,48 b

V159 8,98 b 6,73 a 7,68 a

V161 0,07 b 0,08 b 0,04 a

V162 2,71 b 1,59 a 2,94 b

V165 3,22 b 1,91 a 2,01 a

V166 0,05 b 0,02 a 0,06 b

V167 2,62 a 2,79 a 3,17 a

Tabella 141. Mean separation by multiple range test (Tukey), the comparison is among data shown in

horizontal.


calculated into five new complex variables to represent 94,47% of the total variability of the

aromatic characteristics of the berries at harvest (tab. 142).

The descriptors that represent the highest coefficients (tab. 143) operate in a more reliable

way in determining the aromatic profile of the berries at harvest.

The first component is linked to the total aromatic content of compounds generated by

enzymatic hydrolysis and alcohols monoterpenols classes. Monoterpenols, aliphatic alcohols,

ketones norisoprenoids, oxides monoterpenols, trans 8-OH linalool + cis 8-OH linalool/p-

menth -1-ene-7,8-diol and linalool/geraniol characterize the second component.

The third is linked to 3-hydroxy β-damascenone/3-oxide α-ionol, to trans 8-OH linalool/cis 8-

OH linalool, to hydrocarbon, to monoterpenols and to norisoprenoids. Acids and linalool oxA

/linalool oxB are the only two variables correlated to fourth component.

The last component shows coefficient values below 0,4 (tab. 143).

178

Component


Total

%

variability

%

cumulated 1 8,88 29,59 29,59

2 8,80 29,34 58,93

3 6,33 21,10 80,03

4 2,99 9,99 90,03

5 1,33 4,44 94,47


Variable Component

1 2 3 4 5

Total ,964

Heterosides 1 ,958

Phenols ET1 ,953


Norisoprenoids ET1 ,865 -,466

Vanillins ET1 ,825

Esters ET1 ,789 ,458

V159 -,693 -,449 ,441

Alcohols monoterpenols ET2 -,681 ,615

Monoterpenols ET 1 ,981

V165 ,970

V161 ,934


Oxides monoterpenols ET2 ,852

Aliphatic alcohols ET1 -,766 ,469

V167 ,447

Alcohols + ethers norisoprenoids

ET2 ,975

Heterosides 2 ,896


Hydrocarbon monoterpenols ET2 ,844 -,425

V162 -,575 ,767

V166 ,748

Aldeids ET1 ,740

V158 -,925

Acids ET1 ,803



179

By extracting only two of the components from the total data it is possible to explain 64,32%

of the total variability (tab. 144). The variables examined are subdivided into equal parts

between the two extracted components (tab. 145).


to the ripening index of the berry are directly correlated to it while those further away are

correlated negatively. The first and the second component are negatively correlated only with

the linalool oxC /linalool oxD e linalool/geraniol. The first component is negatively

correlated with the ratios and the monoterpenols, with oxides e alcohols ,with monoterpenols,

and with ketones norisoprenoids. Most aromatic compounds classes and total are, on the

contrary, negatively correlated with the second component (fig. 70).

Component


Total

%

variance

%

cumulated 1 10,59 35,30 35,29

2 8,71 29,03 64,32



Variable Component

1 2

Aldeids ET1 ,853

V 162 ,821 ,516

Vanillis ET1 ,781 -,410

Norisoprenoids ET1 ,746 -,509


Benzene derivates ET1 ,576 -,659

Total ,575 -,635

Heterosides 1 ,563 -,665

Alcohols + ethers norisoprenoids ET2 ,518 ,791

Esters ET1 ,478 -,685

V167

Heterosides 2 ,874


V166 ,745

Acids ET1 -,732

Phenols ET1 -,600

V158 ,547

Aliphatic alcohols ET1

Ketones norisoprenoids ET2 -,533 ,546

180


V159 -,574

Monoterpenols ET 1 -,691

V165 -,812

V161 -,879

Alcohols monoterpenols ET2 -,959



Figure 70. Rotated graph of the first two components obtained from the factorial analysis

Examining the correlation between the parameters analyzed, in the table, only the variables


the value indicated. Esters, linalool oxA /linalool oxB, linalool oxC /linalool oxD, 3-hydroxy

β-damascenone/3-oxyde α-ionol and linalool oxA /linalool oxB, are the parameters that don‟t

present significant Pearson‟s correlations (tab. 146). Pearson‟s significant correlations are

mostly positive except for linalool/geraniol, trans 8-OH linalool/cis 8-OH linalool and

alcohols monoterpenols.

181


Variable

Aliph.

Alc

ET1

Benz.

Der.

ET1

Phen

ET1

Van.

ET1

Mon.

ET 1

Nor.

ET1

Ald.

ET1

Acid

ET1

Hydr.

Monot.

ET2

Oxid

Monot.

ET2

Alc.

Monot

ET2

Hydr.

Noris.

ET2

Ket.

Noris.

ET2

Alc+Et.

Noris.

ET2

Heter.

1

Heter.

2 Tot. V

161

V

162

V

165

Aliph. Alc.

ET1

0,57

Benz. Deriv.

ET1

0,57 0,66 0,78 0,70 0,95 0,94

Phenols ET1 0,66 0,79 0,82 0,60 0,79 0,80

Vanillins ET1 0,78 0,79 0,76 0,88 0,88

Monoterp.

ET 1

0,54 0,53 0,55 0,53 0,64

Norisopr.ET1 0,70 0,82 0,76 0,86 0,87

Aldehyd.ET1 0,60

Acids ET1 0,52

Hydroc.

Monot. ET2

0,67 0,55 0,71 0,70 0,76

Oxid.

Monot. ET2

0,67 0,71 0,74 0,71 0,57 0,76 0,63 0,61

Alcohols

Monot. ET2

0,54 0,55 0,71 0,59 0,63 0,56 0,62

Hydrocarbon

Norisop.ET2

0,71 0,74 0,65 0,86 0,97

Ketones

Norisop.ET2

0,71 0,65 0,63 0,53

Alcoh.+ Eters

Norisop.ET2

0,70 0,57 0,59 0,86 0,93

Heterosides 1 0,95 0,79 0,88 0,53 0,86 0,52

Heterosides 2 0,76 0,76 0,63 0,97 0,63 0,93

Total 0,94 0,80 0,88 0,55 0,87

V161 0,53 0,63 0,56 0,53 0,69 0,84

V162 0,62 0,69

V165 0,64 0,61 0,84

182

The Stepwise discriminant analysis gave the best results to study the characteristics and

possible important statistical differences of the grapes from the theses belonging to the same

Denomination of Origin (fig. 71).


147).

Function Eingenvalue

% of

variance

%

cumulated

Canonical

correlation

1 231,771 83,6 83,6 ,998

2 31,404 11,3 94,9 ,984

3 8,452 3,0 97,9 ,946

4 4,525 1,6 99,6 ,905

5 1,219 ,4 100,0 ,741


Looking at the graph of the centroids obtained from the discriminating analysis data (fig. 71),

is observed that the points relative to centroids present a limited dispersion. In the centroid

there are two distinct groups, the first belonging to a thesis from a different estate from the

other five and it is the only one that stands out. In the second group the other theses: among

these, the second stands out more from the others, while numbers four and five overlap in

some points.

The 100,0% of the original grouping are classified correctly, while 88,7% of the cases


183

Figure 71. Centroids obtained from the cluster analysis of the aromatic characteristics of the grapes at

harvest‟s time.

Thesis

Group expected

Totals 1 2 3 4 5 6

Cro

ss-v

alid

atio

n a

% 1 100,0 ,0 ,0 ,0 ,0 ,0 100,0

2 ,0 100,0 ,0 ,0 ,0 ,0 100,0

3 ,0 ,0 77,8 22,2 ,0 ,0 100,0

4 ,0 ,0 ,0 66,7 33,3 ,0 100,0

5 ,0 ,0 ,0 11,1 88,9 ,0 100,0

6 ,0 ,0 ,0 ,0 ,0 100,0 100,0





184


In the multivariate analysis, if the factor is the thesis, all the variables examined appear to be

statistically important, however, if in the course of the analysis the factor choice is the year

the alcohol aliphatic, the vanillins and the aldehydes derived from the enzymatic hydrolysis

lose their statistical significance. From the interaction thesis by year, some variables are not

statistically different for instance: the benzene derivates, esters, hydrocarbon and

monoterpenols alcohols, ketones norisoprenoids, the total compounds originated by acid

hydrolysis and the content of the identified aromatic compounds expressed in ng/g fresh

weight. The ratios linalool oxC/linalool oxD, linalool/geraniol, trans 8-OH linalool/cis 8-OH

linalool and 3-hydroxi β-damascenone/3-oxo-ionol do not appear statistically relevant only by

the interaction thesis by year (tab. 149).

Factor Dipendent variable

F Sign.

Factor Dipendent variable F Sign.

Year Aliph. alc. ET1 1,077 ,353 Thesis* Year

Aliph. alc. ET1 2,138 ,063

Der.b enzene ET1 11,022 ,000 Der. benzene ET1 1,541 ,185

Phenols ET1 3,823 ,033 Phenols ET1 2,633 ,026

Vanillins ET1 ,232 ,794 Vanillins ET1 1,523 ,191

Monoterp. ET 1 4,390 ,021 Monoterp. ET 1 2,582 ,028

Norisopren. ET1 20,674 ,000 Norisopren. ET1 4,548 ,001

Aldehydes ET1 ,023 ,977 Aldehydes ET1 6,453 ,000

Acids ET1 203,831 ,000 Acids ET1 12,527 ,000

Esters ET1 16,687 ,000 Esters ET1 1,780 ,121

Hydro. monot. ET2 13,380 ,000 Hydro. monot. ET2 1,414 ,231

Oxi. monot. ET2 16,655 ,000 Oxi. monot. ET2 2,493 ,033

Alc. monot. ET2 7,753 ,002 Alc. monot. ET2 1,312 ,276

Hydr. norisopr. ET2 12,937 ,000 Hydr. norisopr. ET2 2,756 ,021

Ket. norisopr. ET2 14,228 ,000 Ket. norisopr. ET2 1,201 ,331

Alc.nor. + Ether ET2 18,947 ,000 Alc. nor. + Ether ET2 2,325 ,045



Total 19,036 ,000 Total 1,425 ,227

V158 3,375 ,048 V158 2,924 ,015

V159 13,964 ,000 V159 1,696 ,140

V161 30,696 ,000 V161 1,373 ,248

V162 44,142 ,000 V162 2,140 ,063

V165 27,556 ,000 V165 3,607 ,005

V166 8,854 ,001 V166 2,162 ,060

V167 4,450 ,020 V167 3,441 ,006

Table 149. Test of the effects between subjects, p= <0,05.

185

Using data previously obtained the amount of variability attributed to the different factor was

calculated (tab. 150). The variability of none of the variables examined is due in percentage to

the interaction thesis by year. The thesis, however, shows more variability as concerns aroma

compounds generated by enzymatic hydrolysis; on the contrary aroma compounds originated

by acid hydrolysis are the variables that show more variability when the factor is represented

by the year. The ratios between the compounds show higher percentage values of variability

for the year.

Variable

%

variability

due to

the thesis

%

variability

due to

the year

%

variability

due to

interaction

thesis/year

%

variability

due to

error

Aliph. alc. ET1 55,29 11,42 22,67 10,61 Derivates benzene ET1 64,01 29,25 4,09 2,65 Phenols ET1 77,26 11,66 8,03 3,05 Vanillins ET1 73,61 2,23 14,59 9,58 Monoterpenols ET 1 69,71 16,68 9,81 3,80 Norisoprenoids ET1 59,35 32,05 7,05 1,55 Aldehydes ET1 84,07 0,05 13,75 2,13 Acids ET1 8,95 85,38 5,25 0,42 Esters ET1 35,95 54,90 5,86 3,29 Hydro. monot. ET2 37,58 52,88 5,59 3,95

Oxi. monot. ET2 43,98 46,31 6,93 2,78

Alc. monot. ET2 42,41 44,37 7,51 5,72

Hydr. Norisopr. ET2 46,56 41,42 8,82 3,20

Ket. norisopr. ET2 24,86 65,07 5,49 4,57

Alc. nor. + Ether ET2 34,06 56,10 6,88 2,96 Heterosides 1 80,88 9,23 7,20 2,69 Heterosides 2 31,24 60,99 4,56 3,20 Total 31,24 60,99 4,56 3,20 V158 82,46 8,11 7,03 2,40 V159 15,48 70,85 8,60 5,07 V161 23,04 71,44 3,20 2,33 V162 30,35 65,03 3,15 1,47 V165 8,89 78,06 10,22 2,83 V166 31,57 50,42 12,31 5,70 V167 28,89 35,59 27,52 8,00


From Tukey test it is possible to note the table with the different subsets and those non

differentiated (tab. 151-153).

186

Within the heterosides 1 classis, only the esters class creates non differentiated homogeneous

subsets while in the heterosides 2 there is not even one. In the compounds released by

enzymes, the benzene derivates family is the most prominent in the theses while the aldehydes

are quantitatively inferior; from a concentration of hundreds of ng/g fresh weight of vegetable

tissue to values close to the unit (tab. 151).

Comparing the theses in the study, the second and third show lower levels of compounds

freed by enzymatic way and the fifth, BM5, stands out for the highest heterosides 1 and 2

content and therefore for total aromas. The grapes from the BM4 sample contain lower

quantities of compounds originated by acid hydrolysis, except for the hydrocarbon

monoterpenols class (tab. 152).

As can be seen from all the ratios examined there is perfect harmony in the results and in the

values obtained in literature by other authors for the „Sangiovese‟ cultivar (Di Stefano et al.,

1998; Lanati et al., 2001). The first and second ratio regarding the linalool oxides are

completely favourable to the trans form compounds compared to the cis one; the

linalool/geraniol ratio is much less than one in the grapes with a low linalool concentration

(tab. 153).

Thesis Aliph.

alcoh.

Der.

benzene Phenols Vanillins Monoter. Norisopr. Aldeids Acids Esters

BM 1 160,58 a 441,89 a 65,19 a 125,95 a 91,09 b 242,10 b 2,2 b 101,77 a 15,60 a

BM 2 140,38 a 437,87 a 59,21 a 120,85 a 66,11 a 165,83 a 2,89 cd 52,98 a 11,04 a

BM 3 151,84 a 471,26 a 52,60 a 111,11 a 75,28 ab 220,42 ab 1,50 a 86,85 a 11,84 a

BM 4 132,60 a 504,47 a 71,27 a 117,05 a 72,52 ab 234,71 ab 2,48 bc 79,96 a 8,45 a

BM 5 268,26 b 815,22 b 115,79 b 213,84 b 147,66 c 463,41 c 3,42 d 162,38 b 13,38 a

Table 151. Significant parameters with different and no differentiated subsets; heterosides 1. Tukey

Test (p=0,05).

Thesis

Hydroc.

monot.

Oxid.

monot.

Alc.

monot.

Hydr.

norisopr.

Ket.

norisopr.

Alc.

norisopr

+ Ethers BM 1 0,36 b 2,86 ab 3,19 a 21,99 ab 6,30 a 19,58 a

BM 2 0,12 a 3,24 b 3,03 a 22,28 ab 6,22 a 15,67 a

BM 3 0,42 b 3,06 b 3,03 a 19,79 a 5,10 a 16,20 a

BM 4 0,24 ab 1,40 a 1,97 a 12,66 a 3,63 a 10,20 a

BM 5 0,72 c 4,13 b 5,59 b 36,15 b 10,63 b 36,62 b


187

Thesis Heter.

1

Heter.

2

Total V158 V159 V161 V162 V165 V166 V167

BM 1 1304,84 a 54,54 a 1359,38 a 1,42 ab 6,77 a 0,08 c 1,63 a 1,95 ab 0,06 ab 4,67 a

BM 2 1098,24 a 50,59 a 1148,86 a 1,59 b 7,89 ab 0,05 b 2,23 b 1,93 ab 0,09 b 2,77 a

BM 3 1225,09 a 47,69 a 1272,77 a 1,47 ab 8,70 b 0,04 ab 2,18 ab 1,40 a 0,06 ab 4,97 a

BM 4 1260,88 a 30,12 a 129,00 a 1,45 ab 8,44 b 0,05 ab 1,74 ab 1,50 a 0,07 ab 2,80 a

BM 5 2277,13 b 93,86 b 2370,99 b 1,33 a 8,07 ab 0,03 a 2,29 b 2,29 b 0,05 a 2,63 a

Table 153. Significant parameters with different subsets; heterosides 1, 2 and ratios. Tukey

Test (p=0,05).

From the multivariate analysis where factor choice is the year and from the Tukey test, it

appears that most of the parameters tested originate homogenous differentiated subsets,

exception made for phenols, norisoprenoids and esters classes. Linalool oxA/linalool oxB

generated by acid hydrolysis is the only ratio between specific compounds that creates

homogenous non differentiated subsets.

The grapes harvested in 2010 stand out for the highest levels of the aromatic families, with the

exception of esters classes which were predominant in 2009. On the contrary, vanillins,

oxides monoterpenols, norisoprenoids, alcohols + ethers monoterpenols, were predominant in

2011 (tab. 154).

Variable 2009 2010 2011

Aliphatic alcohols ET1 181,65 b 192,78 b 137,77 a

Benzene derivates ET1 565,99 b 557,44 ab 478,99 a

Phenols ET1 69,38 a 76,11 a 72,95 a

Vanillins ET1 130,91 ab 126,06 a 156,31 b

Monoterpenols ET 1 79,04 a 112,37 b 80,18 a

Norisoprenoids ET1 265,76 a 267,16 a 262,96 a

Aldehydes ET1 1,47 a 1,54 a 4,49 b

Acids ET1 86,36 a 139,91 b 64,10 a

Esters ET1 14,83 a 12,83 a 8,53 a

Hydroc. monot. ET2 0,17 a 0,44 b 0,51 b

Oxid. monot. ET2 2,04 a 3,35 b 3,43 b

Alc. monot. ET2 2,28 a 4,18 b 3,62 b


Ket. norisopr. ET2 4,19 a 5,69 a 9,25 b

Norisopr. alc.+ Ether ET2 7,52 a 24,10 b 27,34 b

Heterosides 1 1435,53 ab 1557,14 b 1307,02 a

Heterosides 2 26,86 a 63,63 b 75,58 b

Total 1462,39 ab 1620,80 b 1382,61 a

188

Variable 2009 2010 2011

V158 1,28 a 1,58 b 1,49 b

V159 6,67 a 9,94 b 7,31 a

V161 0,07 b 0,06 b 0,02 a

V162 1,66 a 2,06 b 2,34 b

V165 1,62 a 2,23 b 1,59 a

V166 0,06 ab 0,05 a 0,08 b

V167 2,71 a 4,99 a 3,01 a


horizontal.


calculated into six new complex variables (components) to represent 95,08% of the total

variability of the aromatic characteristics of the berries at harvest (tab. 155).

The first component is linked to aromatic compounds released by acid hydrolysis, on the

other hand, the second to classes belonging to the heterosides 1. Monoterpenols, esters and

the most of the variables studied characterize the third component. The fourth and the fifth

component are linked to only one variable precisely trans 8-OH linalool/cis 8-OH linalool e

linalool oxA/linalool oxB respectively. Finally the last component shows coefficient values

below 0,4 (tab. 156).

Component


Total %

variability %

cumulated 1 7,84 26,15 26,15

2 7,09 23,63 49,78

3 6,90 23,01 72,79

4 2,63 8,76 81,55

5 2,40 8,00 89,55

6 1,66 5,53 95,08

Tab 155. Results of the factorial analysis: principal components method.

189

Variable Component

1 2 3 4 5 6 Heterosides 2 ,961


Hydrocarbon monoterpenols ET2 ,956

Alcohols + ethers norisoprenoids ET2 ,936



Oxides monoterpenols ET2 ,624 -,492

Aldehydes ET1 ,595 ,613


Phenols ET1 ,939

Vanillins ET1 ,919



Heterosides 1 ,969

Total ,951

V158 ,784

V159 ,822

V162 ,472 ,834

V165 ,759

V166 -,961

V167 -,954

Acids ET1 -,400 ,513 ,413 ,430

V161 -,490

Esters ET1 -,601

Aliphatic alcohols ET1 -,701 ,482



Extracting only two components from all the data collected it is possible to explain 62,02 %

of the total variability (tab. 157).

In particular the first component is greatly linked to the most of the ratios analysed, to the

classes generated by enzymatic hydrolysis and to ketones norisoprenoids by acid one.

The second, on the other hand, to herterosides 2, to aliphatic alcohols, to monoterpenols and

to acids classes (tab. 158).

The graph (fig. 72) shows how both of the components are negatively correlated with the

esters, the aliphatic alcohols and the linalool/geraniol. The first component is negatively

correlated with the acids and the monoterpenols derived from enzymatic hydrolysis, with

oxides e alcohols monoterpenols from acid hydrolysis, and with the most of ratios between

190

specific aromatic compounds. Benzene derivates, phenols, vanillins, heterosides 1 and 3-

hydroxi β-damascenone/3-oxo-ionol are negatively correlated with the second component.

Component


Total

%

variance

%

cumulated 1 9,70 32,32 32,32

2 8,91 29,70 62,02



Variable Component

1 2

Aldehydes ET1 ,921


Derivates Benzene ET1 ,786

Vanillins ET1 ,781

Total ,734


Phenols ET1 ,706

Heterosides 1 ,691

V162 ,666

V166 -,700

Acids ET1 -,588

V159 -,644 ,706

V158 -,730 ,482

V161 -,851



Esters ET1 -,530

Hydrocarbon monoterpenols ET2 ,866

Oxides monoterpenols ET2 ,838

Alcohols monoterpenols ET2 ,952


Alcohols + ethers norisoprenoids ET2 ,898

Heterosides 2 ,882

V165 ,829

V167



191





Esters, trans 8-OH linalool/cis 8-OH linalool, 3-hydroxi β-damascenone/3-oxo-ionol, linalool

oxA/linalool oxB, are the parameters that don‟t present significant Pearson‟s correlations. In

most cases there are values of Pearson's correlation positive. However Pearson‟s correlations

among ratios and others variables, are negatively correlated and show lower values (tab. 159).

The following variables show Pearson values close to the unit and thus are closely correlated

to each other: aromatic compounds derivated by enzymatic hydrolysis and total aromatic

compounds, the benzene derivates and the norisoprenoids. Those derived from acid hydrolysis

are closely correlated to hydrocarbon norisoprenoids and alcohols + ethers monoterpenols

(tab. 159).

192

Aliph.

Alc

ET1

Benz.

Der.

ET1

Phen

ET1

Van.

ET1

Mon.

ET 1

Nor.

ET1

Ald.

ET1

Acid

ET1

Hydr.

Monot.

ET2

Oxid

Monot.

ET2

Alc.

Monot

ET2

Hydr.

Noris.

ET2

Ket.

Noris.

ET2

Alc+Et.

Noris.

ET2

Heter.

1

Heter.

2 Tot.

V

158

V

159

V

161

V

165

Aliph. Alc.

ET1

0,84

0,69

0,63

0,82

0,72

0,82

0,90

0,89

Benz. Deriv.

ET1

0,84

0,85

0,81

0,74

0,90

0,59

0,96

0,95

Phenols ET1 0,69 0,85 0,82 0,68 0,84 0,88 0,87

Vanillins ET1 0,63 0,81 0,82 0,59 0,85 0,55 0,56 0,83 0,84

Monoterp.

ET 1

0,82

0,74

0,68

0,59

0,75

0,83

0,56

0,68

0,59

0,85

0,87

0,57

Norisopr. ET1 0,72 0,90 0,84 0,85 0,55 0,57 0,93 0,94

Aldehyd.ET1 0,55 0,68 0,59

Acids ET1 0,82 0,59 0,83 0,55 0,74 0,73

Hydroc.

Monot. ET2

0,56

0,57

0,62

0,73

0,79

0,71

0,85

Oxid.

Monot. ET2

0,62

0,86

0,82

0,64

0,74

0,81

Alcohols

Monot. ET2

0,68

0,73

0,86

0,83

0,68

0,84

0,86

Hydrocarbon

Norisop.ET2

0,79

0,82

0,83

0,86

0,94

0,99

Ketones

Norisop.ET2

0,56

0,68

0,71

0,64

0,68

0,86

0,81

0,88

0,56

Alcoh.+ Eters

Norisop.ET2

0,59

0,90

0,74

0,84

0,94

0,81

0,98

Heterosides 1 0,90 0,96 0,88 0,83 0,85 0,93 0,74 1,00

Heterosides 2 0,85 0,81 0,86 0,99 0,88 0,98

Total 0,89 0,95 0,87 0,84 0,87 0,94 0,73 1,00

V158 0,60

V159 0,60

V161 0,59 0,56

V165 0,57


193

The Stepwise discriminant analysis gave the best results to study the characteristics and

possible important statistical differences of the grapes from the theses belonging to the same

estate (fig. 68).

Discriminant analysis on the aromatic characteristics of the grapes at harvest‟s time highlighted

differences between clones examined and some of which may be distinct.


160).

Function

Eingenvalue % of

variance % cumulated

Canonical correlation

1 71,504 79,3 79,3 ,993

2 14,350 15,9 95,2 ,967

3 3,266 3,6 98,9 ,875

4 1,029 1,1 100,0 ,712


In the centroids graph obtained from the discriminant analysis (fig. 73), three distinct groups

appear: the first includes the thesis BM2, BM3 e BM4 that intersect each others. The second

only BM1 and the third BM5 which well differs from the others, because well detached from the

other theses.

The original grouping classified correctly 97,8% of the data, while 84,4% of the cases

grouped cross-validated were reclassified correctly (tab. 161).

194


time.

Cro

ss-

val

idat

oa Thesis

Group expected

Totals 1 2 3 4 5

% 1 100,0 ,0 ,0 ,0 ,0 100,0

2 ,0 77,8 22,2 ,0 ,0 100,0

3 ,0 11,1 66,7 22,2 ,0 100,0

4 ,0 ,0 22,2 77,8 ,0 100,0

5 ,0 ,0 ,0 ,0 100,0 100,0


a. Cross validation is done only for those cases in the analysis In cross validation, each case is

classified by the functions derived from all cases other than that case. b. 97,8% of original grouped cases correctly classified. c. 84,4% of cross-validated grouped cases correctly classified.

195

4. DISCUSSION AND CONCUSIONS


its high responsiveness to the environment so it would be possible to obtain in different areas

wines with quality levels very similar, though differentiated between them, thus expressing

the varietal potential in response to a specific terroir (Brancadoro et al., 2006; Bucelli et al.,

2013; Scalabrelli, 2013). These studies of the wines produced with „Sangiovese‟ as the main

variety in Tuscany mainly identify by Denomination of Origin DOCG and DOC suggest that

possible terroirs identifiable, could be still more numerous (Costantini et al., 2006, 2008;

Scalabrelli, 2013). Research carried out in Tuscany showed that measuring several variables

of vine performance could be used to predict wine quality (Bucelli et al., 2010) although the

grape aroma compound were not determined. This method of testing the wine-making

production allowed us to evaluate phenotypic expression of growing and of quality as a

consequence of interaction between genotype and environment (Scienza et al., 1990; Asselin,

2000; Panont et al., 1994). Generally the product of a genotype isn‟t set strictly but it can

have a broad range of expressions, this range is as extensive as greater is the responsiveness to

local influences.

Although the „Sangiovese‟ grapevine was well studied, it is not possible to indicate a

generalized vineyard model adapt to all situations. Our investigations performed on the grapes

at maturity does not aim to make a hierarchical scale of oenological products of a specific

terroir, but to provide a way to understand the potential of a territory in order to enhance its

specificity.

Our research was conducted in three consecutive years (2009, 2010 and 2011), on

„Sangiovese‟ vineyards in five areas of production located in „Grosseto‟, „Pisa‟, and „Siena‟

provinces, involving a total of 17 theses. The corresponding Denomination areas of wine

production were: „Brunello di Montalcino‟, „Chianti Classico‟, „Chianti Colline Pisane‟,

„Montecucco‟ and „Morellino di Scansano‟: the vineyards are similar as the age, training

system, vine density, bud load and yield were maintained close to the limit imposed by the

denominations (7-8 t/ha).

The examined areas are not homogeneous, as for pedological and climatic characteristics,

these ones are modulated by exposure, altitude and by interaction year by site. It follows that

some grapes presented clearly differences attributable to several factors and in other cases the

effects are difficult to generalize because of phenomena of compensation or of complex

196

interaction among factors. Technical soil and canopy management (shoot trimming and

cluster thinning) were necessary to compensate for the factors that act negatively on quality

parameters of „Sangiovese‟ vines, known to be more sensitive to the ecopedological variables,

compared to other international varieties. In this context, we underline the importance of the

winegrower in order to ensure obtaining a product in conformity with the standards of

production and at the same time having specific quality requirements, and possibly a strong

territorial character (Scalabrelli et al., 2004).

Sometimes the climatic parameters among distant areas are more similar than those ones

among areas very close.

The weather conditions affected significantly several parameter of yield and grape quality

according to the year and the site. In fact, 2010 was the coldest year and 2011, instead, the

hottest in June, August and September determining a generalized advanced of the harvest‟s

time in all the Denomination areas. In 2009, however, the temperature reached in some

locations the highest peaks. Most probably the high temperatures of 2009 brought a

degradation of the aromatic substances which remained the same the following year as a

result of the lower temperatures (D‟Onofrio, oral communications). The maximum, minimum

and mean temperature were negatively correlated with most of the aroma compounds (tab.

162) especially regarding to the compounds released by acid hydrolysis.

Aroma classes

T Max

June-

Sept

T Min

June-

Sept

T Mean

June-

Sept

T Max

Aug-

Sept

T Min

Aug-

Sept

T Mean

Aug-

Sept Monoterpenols

ET 1 P.C. -,605 -,586

Sig. ,004 ,005

Acids P.C. -,537 -,503 -,469

ET1 Sig. ,011 ,017 ,025

Hydroc. monot.

ET2 P.C. -,552 -,585 -,566 -,499

Sig. ,009 ,005 ,007 ,017

Oxide monoterp.

ET2 P.C. -,715 -,658 -,710 -,617

Sig. ,000 ,002 ,000 ,003

Hydroc. norisop.

ET2 P.C. -,672 -,731 -,671 -,599

Sig. ,001 ,000 ,001 ,004

Ketones norisop.

ET2 P.C. -,600 -,581

Sig. ,004 ,006

Alcoh. + Eters

norisop. ET2 P.C. -,428 -,591 -,487 -,537

Sig. ,038 ,005 ,020 ,011

Compounds

released by acid

hydrolysis

P.C. -,636 -,680 -,654 -,588

Sig. ,002 ,001 ,002 ,005

Table 162. Pearson‟s correlations.

197

Maximum, minimum and mean temperature in the period June-July were negatively correlated

with aldehydes and esters, too.

Daily excursion, nevertheless, were negatively correlated with most of the aroma compounds

originating from acid hydrolysis and with terpenols from enzymatic hydrolysis because of the

high peak of maximum temperatures.

Minimun and mean temperature in the period June-September, were negatively correlated

with titratable acidity and positively correlated to ripening technological index. Analysing

only June-July period, were found negative correlation between maximum temperature and skin berry

weight and the total berry polyphenols content.

On the basis of the mean climatic data observed during the three years studied, three distinct

groups were obtained: the first represented by the stations of „San Miniato‟ („Chianti Colline

Pisane‟) and „Magliano‟ („Morellino di Scansano‟) that appear less distinct and by the two

stations belonging to the „Montalcino‟ area. The second is constituted by the station of

„Gaiole‟, and the third by „Cinigiano‟ („Montecucco‟) which is the most distinguishable.

The 2011 was the year with the most mature grapes from a sensory point of view, especially

as regards the pulp. The 2010, instead, showed more seeds phenolic maturity.

Regarding the berry sensorial maturity the lowest values were shown in the thesis of „Chianti

Colline Pisane‟ characterized by modest rainfall, high maximum temperatures, by having silt,

alkaline, calcareous soils, poor of organic matter and potassium and with a mean amount of

magnesium and phosphorus. These characteristics, had a negative effect on concentration of

most of the aroma compounds found in the grapes.

In general, optimal value of sensory maturity were observed in „Siena‟s province and

precisely in the vineyards located in the „Chianti Classico‟: they are characterized by medium

to high content of sand, medium silt and a variable amount of clay. Both soils are alkaline,

moderately rich in active limestone, organic matter and mineral elements. In addition we

observed in this area lower mean temperature during the summer months and slower grape

ripening along two years which assured higher concentration of aroma compounds (tab. 162)

and a better seed phenolic maturity.

Although the training systems utilized, e.g. spur cordon and Guyot cannot be compared, it

was observed a tendency on the grapes brought by the last pruning system to induce lower

values of sensorial maturity of berries.

In the year 2009 grapes achieved the highest values of the Ripening Index parameter (sugar

content/titratable acidity) while on 2011 it was observed generally a better phenolic maturity.

198

Regarding the sensorial maturity the values shown are the highest in one thesis of the „Chianti

Classico‟ area and the lowest in the thesis of „Chianti Colline Pisane‟, which showed the

lower seed phenolic maturity while a thesis belonging to „Brunello di Montalcino‟ showed the

ripening technological index highest and the highest value of total polyphenols.

As it was not found any correlation between the amount of aromas perceived in sensory

analysis with aroma analysed by the GC-MS; this fact could be depending on several reasons:

the non contemporary sensorial analysis between the tested thesis, the different threshold of

perception of the several aroma compounds, (De Rosso et al., 2010) and the saturation of

perception of the panelist which occurs when the threshold was overcame.

As for sensory analysis of the grapes and for the technological parameters, the effect due to

the year appeared much more relevant than those dependent on the site.

It was found that the mean bunch weight was positively correlated to the nitrogen

concentration of the soil.

Sugar and polyphenols content showed the highest values in theses characterized by good

daily excursion, an average rainfall able to increase the available water content of the soil

(AWC) and soils for the prevalence of silty particles, alkaline pH, medium quantity of active

limestone, poor of organic matter and of other macro elements (Brancadoro et al., 2006).

The lowest °Brix and the highest value of titratable acidity, on the other hand, were found in

soils characterized by medium to high content of sand, medium silt and variable amount of

clay; alkaline, moderately rich in active limestone, organic matter and mineral elements;

altogether the annual thermal regime can play an important role on sugar accumulation and

final titratable acidity. A direct positive correlation was found between sugar accumulation

and potassium content of the soil.

Soils characterized by a prevalence of sand, sub-alkaline pH, moderate presence of limestone,

low organic matter, and average potassium, magnesium and exchange cation capacity

positively influenced anthocyanins level as in most vineyards of „Montecucco‟ area; on the

contrary they were negatively conditioned by soils clay-sandy (with a consistent content of

clay), poor in organic matter and phosphorus, rich in potassium and with an excess of

magnesium (BM 6).

With regard to the aroma profile of „Sangiovese‟‟s grapes, we can observe how the results

here reported for all the theses examined are in perfect agreement between them and with the

general profiles obtained by other authors for the „Sangiovese‟ (Di Stefano et al., l.c.; Lanati

et al., l.c.). Specifically typical ratios of „Sangiovese‟ aroma compounds are linalool

oxA/linalool oxB, linalool oxC/linalool oxD, linalool/α-terpineol, linalool/geraniol, trans 8-

199

OH linalool/cis 8-OH linalool, trans 8-OH linalool + cis 8-OH linalool/p- menth-1en-7,8-di

and 3-hydroxy β-damascenone/3-oxide α-ionol.

Inside of the compounds released by enzymes hydrolysis, the classes of benzene derivates and

norisoprenoids are present in the most quantity in the samples analyses, while aldehydes are

quantitatively less represented. Chemical hydrolysis to pH 3 made on aglycones obtained by

enzymatic hydrolysis of glycosides precursors has shown that the most represented category

is that of the norisoprenoids hydrocarbon while that of monoterpenols is the less abundant.

Comparing the grapes coming from the different locations we can observe how the „Chianti

Colline Pisane‟ thesis is relatively distinct from the others for the quantitative inferior values

of most aroma classes, which can be attributed to the lower soil content of potassium in this

vineyard. In fact, a positive correlation (r2 = 0,631) with the aroma concentration with soil

potassium content was found.

Samples from the province of „Siena‟, however, had in general a high aroma content. In

particular only one thesis belonging to „Brunello di Montalcino‟ mentioned for the highest

ripening technological index and the value of total polyphenols, besides showed the highest

aroma‟s content.

Among the soil studied, one thesis in the „Chianti Classico‟ area stood out while the other

ones are close to one another and some points overlap: the soil of this vineyard is sub-

alkaline, with prevalence of clay and it is characterized by the highest altitude.

Besides, a good daily excursion, a medium rainfall, a soil alkaline, with the prevalence of silty

particles, with a fair quantity of active limestone but poor of organic matter and of other

macro elements, positively affect the aroma‟s content. Moreover the percentage of the sand in

the soils appeared positively correlated to the level of hydrocarbon, alcohols and ethers

norisoprenoids derivates.

The year 2010 was characterized by the highest level of the most aroma classes while on 2009

the final concentrations were lower, as reported before. Variability was equally attributable to

the thesis and to the year more precisely, the compounds, originating from enzymatic

hydrolysis except for aldehydes and acids, have a greater variability due to the thesis, while

the compounds originating from acid hydrolysis and the other ratios are strongly influenced

by the year effect. Most of the compounds derived by acid hydrolysis and some derived by

enzymatic hydrolysis were negatively correlated to the temperature occurred in the period of

berry growth and ripening (from the beginning of June to the end of September). Moreover,

sugar content and the ripening technological index were negatively correlated with most of

the aroma compounds, while titratable acidity and mean bunch weight showed a positive

200

correlation. Only the class of acids obtained from enzymatic hydrolysis was positive

correlated with °Brix.

Berry size was correlated negatively with anthocyanins and polyphenols concentration while

was not correlated with the aroma content, also it can be highlighted that a positive

relationship with the content of limestone and aroma compounds, referred in literature

(Fregoni, 2005) was not evidenced.

Regarding to the focus of the vineyards of „Montecucco‟ area there were several differences

among them: soil of the vineyard MC7 has a lower pH than the others, almost no limestone,

low organic matter and available nitrogen, meanwhile it has a good amount in phosphorus,

magnesium and potassium content. This thesis which differs from other ones as regards the

characteristics of the soil, not for the climate, because is very close, showed different results

of sensorial, technological and aroma analysis, more provided mainly of norisoprenoids,

phenols and monoterpenols. These differences were partially accounted by the high content of

sand and potassium as resulted by the positive correlation previously indicated.

As for the clone effect grown in the same site of cultivation („Col d‟Orcia‟ estate), three

distinct groups appeared during technological and aromatic analysis: the first comprises the

thesis BM2, BM3 e BM4 that intersect each others. The second only BM1 and the third BM5

which well differs from the others. The five theses appear well detached when they were

examined for sensorial description. It was confirmed that genetic variability affect the

characteristics of „Sangiovese‟ (Egger et al. 2001; Bertuccioli, 2006) although it was less

relevant than the year and the site.

Secondary metabolites are so sensitive to environmental and cultural variables as can be

synthesized in very different levels depending on the influence of the cultivation conditions,

while preserving their quality profile essentially unchanged. It seems that rainfall didn‟t

generate significant correlation with variables linked to the characteristics of the grape.

Some aroma classes (aliphatic alcohols, benzene derivates, phenols), especially as regards the

compounds released by enzymatic hydrolysis, didn‟t correlated with climatic characteristics.

We should suppose that the biosynthesis of these compounds in „Sangiovese‟ are less

influenced by the climatic conditions. It also could be hypothesized that in the area of

cultivation tested even though we had variable conditions, there were not reached limiting

conditions able to modify their biosynthesis.

The main hypothesis of this study was that soil and climate might have a direct relation upon

grape characteristics but there were many consistent grapes effects that could be linked to

both site and climatic characteristics. So, as already highlighted by Reynolds (2012), if we

201

would have studied also the wines, probably we could have generated several questions: „what

factors exert the greatest control over the terroir effect?; do winemakers exert more influence

over wines than site characteristics?; do viticultural and/or oenological practices exert the

greatest influence over wine varietal typicity?‟

Questions which suggest to continue to study the „Sangiovese‟ in Tuscany, a place in which is

able to give several expressions.

202

5. REFERENCES

A.A.V.V. (1996). Giornata di studio sul Sangiovese. Atti Arezzo: 55-63.

A.A.V.V. (2008). Vino e Olio in Toscana. Ed. Il Fiore, Firenze.

ARYANA P., WILSON B., STRAUSS C. R., WILLIAMS P. J. (1987). The properties of

glycosidases of vitis vinifera and a comparison of their β-glucosidase activity with that of

exogenous enzymes. An assessment of possible applications in enology. Am. J. Enol. Vitic.

38(3): 182–188.

ARTEA (2008). Il potenziale viticolo della Toscana. Newsletter, n. 31, marzo: 1-30.

ASSELIN C. (1999). Territoires et zones viticoles: aspect climatiques, pedologiques,

agronomiques. Caractérisassion des terroirs viticoles: un ètude systémique. Atti Simp. Int.

„Territorio e vino: Siena‟, Ed. Cantagalli-Città del vino, Siena: 57-70.

ASSELIN C. (2001). Relazioni e interazioni tra i fattori dell’ambiente viticolo e la qualità

dell’uva e dei vini. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA: 151-158.

ASSELIN C., MORLAT R., BARBEAU G. (2003). Les Terroirs Viticoles: du concept au

produit. In „Terroir Zonazione Viticoltura‟, Phytoline Editore, Rivoli Veronese: 159-186.

AUSTERWEIL G. (1946)-Ind. Parfum: 1-195.

BANDINELLI R., LORETI F., MANCUSO S., PISANI P.L., PORCINAI S.,

SCALABRELLI G. (2001). Osservazioni comparative su sette cloni di ‘Sangiovese’ nel

‘Chianti Classico’. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA: 219-232.

BANDINELLI R., BOSELLI M., DI COLLALTO G., PISANI P.L. (2006). Ricerche

Preliminari su alcuni biotipi di Sangiovese reperiti in Toscana. Atti Simp. Int. „Il Sangiovese

identità e peculiarità di un vitigno tipico e internazionale‟ Firenze, Ed. ARSIA: 297-306.

BAGNOLI G., BRUGALI L., DI COLLALTO G., SCALABRELLI G. (2001). Ricerche sulla

densità di piantagione del ‘Sangiovese’ in Maremma toscana: risultati di un triennio di

osservazioni. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA: 271-276.

BAGNOLI R., MATTII G.B., GALLUZZI G. (2006). Valutazione bio-agronomica di cloni di

Sangiovese in Toscana. Atti Simp. Int. „Il Sangiovese‟, identità e peculiarità di un vitigno

tipico e internazionale‟ Firenze, Ed. ARSIA: 371-378.

BASSO M. (1982). ‘Agrumi, frutta e uve nella Firenze di Bartolomeo Bimbi pittore mediceo’.

F.&F. Parretti Grafiche, Firenze: 137-157.

203

BAYANOVE C.(1993) Les composés terpeniques.: Les acquisitions récentes en

chromatographie du vin. Applications á l´ analyse sensorielle des vins, B. Doneche, (Ed.), 99-

119, ISBN: 2877773647, Paris, France .

BERTUCCIOLI M., FERRARI S., SILIANI A., ZINI S. (2001). Il ruolo del Sangiovese per

la produzione di vini di qualità. Atti Simp. Int. „Il Sangiovese‟ Firenze, 15-17 febbraio 2000.

Ed. ARSIA: 364-372.

BERTUCCIOLI M., FERRARI S., LENCIONI L., SILIANI A., ZINI S. (2001). Effetti della

densità di piantagione e del sesto di impianto sulla composizione delle uve e vini di

Sangiovese in due zone del Chianti Classico. Atti Simp. Int. „Il Sangiovese‟ Firenze ,15-17

febbraio 2000. Ed. ARSIA: 457-470.

BERTUCCIOLI M. (2006). Idoneità e peculiarità del Sangiovese e gusto internazionale. Atti

Simp. Int. „Il Sangiovese, identità e peculiarità di un vitigno tipico e internazionale‟ Firenze,

Ed. ARSIA: 253-264.

BIAGIOLI G. (1999). Les vignes en pente des ‘Cinque terre’ une merveille humane en

danger, in La protection des terroir viticoles. CERVIM Univ M de Montaigne Bordeaux III,

Paris: 114-222.

BOGONI M. (1998). Capire il territorio per attuare una politica di qualità: I risultati della

zonazione viticola del territorio di Castagneto Carducci. Il Consenso (1-2).

BOGONI M. (1999). La zonazione della DOC Bolgheri: aspetti metodologici ed applicativi.

Atti del Simp. Int. „Territorio e vino: Siena‟, Ed. Cantagalli-Città del vino, Siena: 1877-198.

BOIDRON J. N. (1978)-Ann. Technol. Agric. 27 (1) 141.

BOSELLI M., PISANI P.L., BANDINELLI R., MANCUSO S., MASI E. (2001). Influenze

genetiche e colturali sull’equilibrio produttivo del vitigno Sangiovese. Atti Simp. Int. „Il

Sangiovese‟ Firenze, Ed. ARSIA: 213-218.

BOSELLI M. (2006). Tradizione e innovazione nella gestione tecnico-viticola del

Sangiovese. Atti Simp. Int. „Il Sangiovese, identità e peculiarità di un vitigno tipico e

internazionale‟ Firenze, Ed. ARSIA: 171-175.

BOSELLI M., BANDINELLI R., PISANI L., LORETI F., SCALABRELLI G, TRIOLO E.,

MATERAZZI A., BERTUCCIOLI M., PORCINAI S. (2006). Risultati vegeto produttivi di

sette nuovi cloni di ‘Sangiovese’ nel territorio di produzione del Chianti Classico. Atti Simp.

Int. „Il Sangiovese, identità e peculiarità di un vitigno tipico e internazionale‟ Firenze, Ed.

ARSIA: 523-528.

204

BRANCADORO L., FAILLA O., SCIENZA A. (2006). Indagini sulla maturazione fenolica

del Sangiovese nell’ambiente di Montalcino. Atti Simp. Int. Il „Sangiovese, identità e

peculiarità di un vitigno tipico e internazionale‟ Firenze, Ed. ARSIA: 361-364.

BRANCADORO L., FAILLA O., TONINATO L., SCIENZA A. (2006). Risultati di indagini

di zonazione a diversa scala effettuate nel territorio toscano. Quad. Vitic. Enol. Univ. Torino,

28, 2005-2006: 281-291.

BRANCADORO L.,TONINATO L., FAILLA O., MARIANI L., SCIENZA A. (2006).

Influenza di diversi pedoclimi toscani sulle risposte produttive e qualitative del Sangiovese.

Atti Simp. Int. „Il Sangiovese, identità e peculiarità di un vitigno tipico e internazionale‟

Firenze, Ed. ARSIA: 353-360.

BUCELLI P., COSTANTINI A.C., STORCHI P. (2010). It is possible to predict Sangiovese

wine quality through limited number of variables measured on the vines. J. Int. Sci. Vigne

Vin, 44, n°4: 207-218.

CABRITA J.M.,FREITAS C.A.M., LAUEANO O., DI STEFANO R. (2006). Glycoside

aroma compounds of some Portuguese grape cultivar–J. Sci. Food Agric. 86: 926-931.

CAMPOSTRINI F., COSTANTINI E. (1996). Vino nobile di Montepulciano: zonazione e

valorizzazione del territorio. Ed. Regione Toscana, Firenze.

CARBONNEAU A. (1993). Définition du terroir ou des terroirs. Comm. 8° session G. E.

Physiologie de la vigne, O.I.V., Paris.

CARBONNEAU A. (2000). Climat et sol: critères d’évaluation et effets sur le comportement

de la vigne. 3° Simp. Int. Zonificacion vitivinicola, 6-13 de Mayo 2000, Tenerife, Sotés ed.,

Madrid.

CARBONNEAU A. (2003). Ecophysiologie de la vigne et terroir. “Terroir Zonazione

Viticoltura”, Phytoline Ed., Rivoli Veronese: 61-102.

CASTERAN P. (1971). Conduite de la vigne. Sciences et techniques de la vigne- Tome II‟,

Dunod, Paris : 101-207.

CORDONNIER R. (1956) - Recherches sur l’aromatisation et le parfum des vins doux

naturels et des vins de liqueur. Ann. Technol. Agr. 5, 75-110.

CORDONNIE R. e BAYANOVE C. (1974) C. R. Acad. Sci. Paris 278- 387.

COSTANTINI E.A.C., CAMPOSTRINI F., ARCARA P.G., CHERUBINI P., STORCHI P.

and PIERUCCI M. (1996). Soil and climate functional characters of grape ripening and wine

quality of „Vino Nobile di Montepulciano‟. Acta Hort. 427 ISHS: 45-55.

COSTANTINI E.A.C., BARBETTI R., L‟ABATAE G., BUCELLI P., PELLEGRINI S.,

STORCHI P. (2006). La cartografia della provincia di Siena attraverso un approccio

205

climatico, geomorfologico, pedologico ed agronomico. Atti Conv. Int. Paesaggi, terroirs e i

paesaggi del vino. Perugia 6-9 novembre 2006. AIS: 123-139.

COSTANTINI E.A.C., BARBETTI R., BUCELLI P., L‟ ABATE G., LELLI L.,

PELLEGRINI S., STORCHI P. (2006). Land peculiarity of the vine cultivation areas in the

province of Siena (Italy), with reference to the viticultural and oenological results of

Sangiovese vine. Boll. Soc. Geol. It., Vol. Speciale n. 6: 147-159.

COSTANTINI E.A.C., PELLEGRINI S., VIGNOZZI N. and BARBETTI R. (2006).

Micomorphological characterization and monitoring of internal drainage in soils of

vineyards and olive groves in central Italy. Geoderma 131: 388-403.

COSTANTINI E.A.C., BUCELLI P. (2008). Suolo ed altre colture di qualità: l’introduzione

e la pratica dei concetti di ‘terroir’ e zonazione. Ital. J. Agron./Riv. Agron., 1 Suppl.: 23-33.

COSTANTINI E. A.C., BARBETTI R., BUCELLI P., L‟ABATE G., PELLEGRINI S.,

STORCHI P. (2008). Scale dependance of soil and climate functional characteristics for

qualitative Sangiovese vine production. Atti Congresso Mondiale della vigna e del vino OIV,

Verona,15-20 giugno 2008: 1- 8.

COSTANTINI E.A.C., PELLEGRINI S., BUCELLI P.,STORCHI P., VIGNOZZI N.,

BARBETTI R., CAMPAGNOLO S. (2008). Using ΔC13 to assess viticultural and

oenological suitability for Sangiovese of different pedoclimatic conditions in Chianti. VIIth

Int. terroir Congress, Nyon (Swiss) 19-23 mai 2008 ACW:1-6.

COSTANTINI E.A.C., BUCELLI P., STORCHI P.,PELLEGRINI S., VIGNOZZI N.,

PRIORI S., BARBETTI R., NATAREKKI L., PERRIA R., L‟ABATE G. (2010) Carattteri e

qualità pedologiche e climatiche funzionali alla valorizzazione della vocazione del territorio

al Sangiovese Riv . Enol.N 1-2-3-4 149-161.

COTARELLA R. (2001). Aspetti del marketing del Sangiovese. Atti Simp. Int. „Il

Sangiovese‟, Firenze, Ed. ARSIA: 63-66.

CRICCO J. and TONINATO L. (2004). La zonazione di Cerreto Guidi e Vinci. La scoperta

di un territorio. Comuni di Cerreto Guidi e Vinci.

DELOIRE, A. LOPEZ, F. and CARBONNEAU, A. (2002). Responses de la vigne et terroir

elements pour une methode d'etude. Progrès Agricole et Viticole 119(4): 78-86.

DE ROSSO M., PANIGHEL A. (2010). Chemical Characterization and Ecological Potential

of Raboso Varities by study of Secondary Grape Metabolites J.Agric. Food Chem 11364-

11371.

DI COLLALTO G., SCALABRELLI G. and FERRONI G. (2001). Osservazioni preliminari

sull’attività vegeto-produttiva del ‘Sangiovese’ innestato su tredici portinnesti nella zona di

206

produzione del Morellino di Scansano. Atti Simp. Int. „Il Sangiovese‟ Firenze , Ed. ARSIA:

259-264.

DI STEFANO R.,BOTTERO S., PIGELLA R., BORSA D., BEZZO G., CORINO L. (1998).

Precursori d’aroma glicosilati presenti nelle uve di alcune cultivar a frutto colorato.

L‟enotecnico: 63-73.

DI STEFANO R., GENTILINI N.,UMMARINO I. (2000). Studio dei profili aromatici di

varietà a frutto bianco coltivate nella zona del collio. L‟enologo: 95-102.

DI VECCHI STARAZ M., BANDINELLI R., BOSELLI M., PATRICE T., BOURSIQUOT

J.M., LAUCOU V., LACOMBE T. (2007). Genetic Structuring and Parentage Analysis for

Evolutionary Studies in Grapevine: Kin Group and Origin of the Cultivar Sangiovese

Revealed. Journal of the American Society for Horticultural Science.

D‟ONOFRIO C. (2011). Caratterizzazione funzionale di alcuni geni della biosintesi degli

aromi delle uve durante lo sviluppo dell’acino e controllo della qualità aromatica delle uve.

Italus Hortus 18 (1): 39-61. Review n.13.

D‟ONOFRIO C., MATARESE F., SCALABRELLI G., BOSS P.K (2012). Caratterizzazione

funzionale di alcuni geni della biosintesi degli aromi nelle uve. Italus Hortus: 3-5, 68-71.

DUCCI E., D‟ONOFRIO C., FERRONI G., SCALABRELLI G. (2012). Contributo

dell’analisi sensoriale allo studio della maturità delle uve ‘Sangiovese’ di diverse zone della

Toscana. Acta Italus Hortus :3-5, 144-149.

EGGER, E. GRECO, G. STORCHI, P., GRILLO, S. (1996). First results of the research on

the ecology of the Sangiovese variety in the province of Arezzo. Acta Hort. (427) : 295-302.

ENZEL C. R. (1985)–Pure Appl.–Chem. 57 (5) 693.

FAILLA O., SCIENZA A. (1999). Le zonazioni, esercizio accademico e servizio al mondo

vitivinicolo?. Supplemento a L‟informatore agrario16:5.

FALCETTI M. 1994. Le terroir. Qu’est qu’un terroir? Pourquoi l’etudier? Pourquoi

l’enseigner?. Bulletin de l‟O.I.V., 757-758: 246-275.

FREGONI M. (2003). Definizione di terroir e di zonazione. ‘Terroir zonazione viticoltura’

.Phytoline editore, Rivoli veronese: 23-25, 115-128.

FREGONI M. (2005). Viticoltura di Qualità. Phytoline.

FREGONI M. (2006). Il Sangiovese, vitigno autoctono e internazionale: evoluzione e

strategie di sviluppo. Atti Simp. Int. „Il Sangiovese. identità e peculiarità di un vitigno tipico e


207

GALLENTI G., COSMINA M. (2001). L’analisi sensoriale e la valorizzazione qualitativa

del prodotto vitivinicolo: problematiche teoriche e verifiche empiriche. Atti Simp. Int. Il

„Sangiovese‟. Firenze, Ed. ARSIA: 67-78.

GIANNETTI F., FAVIERE V., BUCELLI P., PIRACCI A.( 2001). Caratteri enologici di

Sangiovese in quattro diverse zone della Toscana. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed.

ARSIA: 379-384.

GROSSMAN M., SUKRAN I. RAPP A. (1990). Deutsche Lebensmitt. Rund

GUNATA Y. Z. (1984). Rechercher sur la fraction lièe de l’arome du raisin. Importance des

terpenylglycosides. Action des glycosidases. These Dr. Ing. University of Montpellier, France.

GUNATA Z., BIRON C., SAPIS J. C., BAYANOVE C. (1989).Vitis 28 191.

GUTH H. (1997). Quantification and sensory studies of character impact odorants of

different white wine varieties. J.Agric.Food Chem. 45: 3027-3032.

IACONO, F., PORRO D., BRANCADORO L., SCIENZA A. (2001). Tipicizzazione del vino

Sangiovese attraverso tecniche di assemblaggio di differenti selezioni clonali: applicazione

innovativa dell’analisi sensoriale. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA: 461-

468 .

INTRIERI C. (1995). La densità di impianto e l’equilibrio fisiologico: due principi a

confronto per la viticoltura del 2000. Rivista di Frutticoltura, 11: 11-23.

INTRIERI C., PONI S. (2000). Physiological response of winegrape to management practice

for successful mechanization of quality vineyards. Acta Hort. (ISHS) 526: 33-48.

JOHNSON H., HALLIDAY J. (1992). The sinther’s Art. New York, Simmer and Shuster.

Scienza A. 1998. La zonazione tra natura e cultura. Il Consenso (1-2): 31-38.

KALUA C.M.,BOSS P.K., (2009). Evolution of voltaic compounds during the development

of Cabernet- Sauvignon grapes. J. Sci. Food Agric.

LANATI D., BONONO M., MAZZA G. (2001). Precursori d’aroma e profilo varietale della

cultivar Sangiovese. Atti Simp. Int. Firenze 15-17 febbraio.447- 452.

LAWLEY F. (1870). Modo di coltivare le viti e di fare il vino. Manuale del Vignajuolo. Ed.

Bettini, Firenze.

LORETI F., SCALABRELLI G. (2007). Sistemi di impianto, forme di allevamento, tecniche

di potatura. In „Storia Regionale della Vite e del Vino in Italia. Toscana‟, Firenze: 393-444.

LOSCOS N., HERNANDEZ HORTE P., CACHO J., FERREIRA V. (2007). Release and

formation on varietas aroma compounds during alcoholic fermentation from non floral grape

odorless flavor precursors fraction. J. Sci. Agric. Food Chem. 55: 6674-6684.

208

MAINARDI G. (2000). Storia di un grande protagonista dell’enologia italiana: il Sangiovese

Atti Simp. Int. Firenze, Palazzo dei Congressi Arsia 15- 17 Febbraio 2000, 17-25.

MALTMAN A. (2008). The role of Vineyard Geology in Wine Typicity Journal of Wine

Research Vol. 19 No 1: 1-17.

MARAIS J. (1983). Terpenes in the aroma of grapes and wines, a review S. Afr J. Enol vitic

4: 49-60.

MATEO J.J., GENTILINI N., HUERTA T., JIMENEZ M., DI STEFANO R. (1997).

Fractionation of glycoside precursors of aroma in grapes and wine. Ed. Elsevier : 219-224.

MATTII G.B., ORLANDINI S., CALABRESE C.E. (2005). Analysis of grapevine vegeto-

productive responses to plant density and rootstock. XIV Int. GESCO Viticulture Congress,

Geisenheim, Germany, 23-27 August, 2005: pag. 6.

MORETTI G., SCALABRELLI G. (2007). Catalogo dei Cloni di vite Omologati dei vitigni

ad uva da vino in Italia. Unione Italiana Vini – Enovitis: 1-207.

MORLAT R. (1989). Le terroir viticole: contribution a l’ètude de sa carecterisation et de son

influence sur le vins; application aux vignolese rouges de la Moyenne Vallee de la Loire.

These d‟Etat Bordeaux II.

MORLAT R., GUILBAULT P., THELIER-HUCHE L., RIOUX D. (1999). Ètude intégrée et

alleégée des terroirs viticoles en Anjou: Caractérisation et zonage de l’unité terroir de base,

en relation avec une enquête parcellaire. Atti del Simposio Internazionale „Territorio e vino:

Siena‟, Edizioni Cantagalli-Città del vino, Siena: 293-306.

MORLAT R. (2001). Terroirs viticoles: ètude et valorisation. Openplurumedia Chaintrè.

ORLANDINI S., DELLA MARTA A., ZIPOLI G., MANCINI M. (2003). Monitoraggio

agroclimatico a scala aziendale. Supplemento a „L‟informatore agrario‟, 14: 19-25.

PALLIOTTI A., CARTECHINI A., PETOUMENOU D., STORCHI P., VALRENTINI P.,

DEI BANCHI R., PERSIA A. (2001). Valutazione dell’attività vegetativa e produttiva del

Sangiovese innestato su diversi portinnesti in Italia centrale. Atti Simp. Int. „Il Sangiovese

identità e peculiarità di un vitigno tipico e internazionale‟. Firenze, Ed. ARSIA: 379-386.

PINNA M.B., MURRU M., ARGIOLAS F., ARGIOLAS A., FILIPPINO G., FLORIS T.,

SCRUGLI S. (2009). Aromatic characterizazion of Cagliari Malvasia. III Simposio Inter.

Malvasia Spagna 26-30 maggio.

PIRACCI A. (2001). Innovazioni tecnologiche per l’esaltazione delle caratteristiche

specifiche del Sangiovese. Atti Simp. Int. „Il Sangiovese‟ Firenze, 15-17 febbraio 2000. Ed.

ARSIA: 373-378.

209

PISANI P.L., BOSELLI M., BANDINELLI R., LORETI F., SCALABRELLI G. (2000).

Risultati conseguiti dalle ricerche nel quinquennio 1995-1999: aspetti agronomici, biologici,

fisiologici ed ampelografici. Progetto Chianti Classico 2000 settore Viticoltura, 5: 11-68.

PONI S. (2001). Fisiologia di comportamento del Sangiovese: aspetti di base e

considerazioni applicative. Atti Simp. Int. „Il Sangiovese‟ Firenze, 15-17 febbraio 2000. Ed.

ARSIA: 159-167.

PONI S. (2003). La potatura verde nel vigneto aspetti fisiologici e colturali. L‟Informatore

Agrario (26), 37-49.

PONI S., BERNIZZONI F.,CIVARDI S. (2008). The effect of early leaf removal on whole

canopy gas exchange and vine performance of Vitis vinifera L. ‘Sangiovese’ Vitis 47 :1-6.

QUADERNI TECNICI VCR (2003). I cloni originali vcr di Sangiovese 1-27.

RAPP A. (1987). In frontiers of flavor. Ed. Charalambous Ensevier Amsterdam.

REMORINI D., SCALABRELLI G., MASSAI R., TATTINI M. (2007). Risposte fisiologiche

di due vitigni al diverso regime idrico, con particolare riguardo al contenuto di sostanze

fenoliche nelle foglie e negli acini. Italus Hortus, vol. 14 (3): 382-385.

REYNOLDS G. (2012). Soil and Climate Interaction with Grapevines. IXe International

Terroirs Congress 2012.

RIBEREAU-GAYON P., DUBOURDIEU D.,DONECHE B., LONVAUD A. (2003).

Trattato di Enologia 1. Microbiologia del vino e Vinificazioni. Edagricole, Bologna.

RIBEREAU-GAYON P., DUBOURDIEU D.,DONECHE B., LONVAUD A. (2003).

Trattato di Enologia 2. Chimica del vino, stabilizzazione e trattamenti. Edagricole, Bologna.

RIBEREAU-GAYON P., BOIDRON,J.N. & TERRIER,A. (1975). Aroma of muscat grape

varieties. J.Agric.Food Chem. 23(6):1042-1047.

RIOU C., MORLAT R., ASSELIN C. (1995). Une approche intégrée des terroirs viticoles:

discussion sur les critères de caractérisation accessibles. „Bulletin O.I. V.‟ 767-768: 92-106.

RYONA I., PAN B.S., INTRIGLIOLO D.S.,LAKSO A.N., SACKS G.L. (2008). Effects of

cluste light exposure on 3-isobutyl-2-methoxypyrazine accumulation and degradation pattern

in red wine grape .J. Agric. Food Chem.

SCALABRELLI G., DI COLLALTO G. (1999). Progetto per il rinnovamento e la

valorizzazione della viticoltura della provincia di Grosseto. In „Rinnovamento e

Valorizzazione della viticoltura della provincia di Grosseto‟. Provincia di Grosseto. ARSIA.

Regione Toscana: 1-64.

210

SCALABRELLI G., LORETI F., PISANI P.L., BANDINELLI R., PORCINAI S. (2001).

Effetto della densità di piantagione sull’attività vegeto produttiva del ‘Sangiovese’ nel

Chianti Classico. Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA: 233-240.

SCALABRELLI G., LORETI F., PISANI P.L., BANDINELLI R. and PORCINAI S. (2001).

Comportamento bio-agronomico del ‘Sangiovese’ su diversi portinnesti nel Chianti Classico.

Atti Simp. Int. „Il Sangiovese‟ Firenze, Ed ARSIA: 233-240.

SCALABRELLI G. (2002). La Vitivinicoltura e il carattere del ‘Morellino’. In Ciuffoletti Z.,

Nanni P. Un vino di Maremma: il Morellino di Scansano. Editrice Laurum: 77-129.

SCALABRELLI, G. FERRONI G., D'ONOFRIO C., DI COLLALTO G., VENERINI F.

(2003). Trials with ‘Sangiovese’ grafted on several grapevine rootstocks in two different

areas of Tuscany. Acta Hort. (ISHS) 617:73-83.

SCALABRELLI G., BORGO M., D‟ONOFRIO C., ANGELINI E. and DE LORENZIS G.,

(2006). Il recupero del germoplasma viticolo nell’area della DOC Montecucco. Convegno „I

Vitigni Autoctoni Minori‟, Villa Gualino, Torino, 30-11-2006/01-12-2006. Vol. unico,

Sessione 3 N-Z: 6-13.

SCALABRELLI G., D‟ONOFRIO C., MACCHIA A., ROSELLINI D., BERTUCCIOLI M.,

BOSELLI M. (2006). Comportamento agronomico del vitigno ‘Sangiovese’ in undici zone del

Chianti Classico. Atti Simp. Int. „Il Sangiovese identità e peculiarità di un vitigno tipico e


SCALABRELLI G., D‟ONOFRIO C., FERRONI G., BAGNOLI G., DI COLLALTO G.

(2006). Osservazioni sul comportamento del vitigno ‘Sangiovese’ allevato a diversa altezza

della parete vegetativa. II° Simp. Int. „Il Sangiovese identità e peculiarità di un vitigno tipico

e internazionale‟ Firenze, Ed ARSIA: 205-214.

SCALABRELLI G., D‟ONOFRIO C., FERRONI G., BAGNOLI G., DI COLLALTO G.

(2006). Osservazioni sul comportamento del vitigno ‘Sangiovese’ allevato a diversa altezza

della parete vegetativa. Atti Simp. Int. “Il Sangiovese identità e peculiarità di un vitigno

tipico e internazionale” Firenze, Ed. ARSIA: 205-214.

SCALABRELLI G. (2007). I vitigni autoctoni.Locus (n. 7), Felici Editore, Pisa: 45-52.

SCALABRELLI G. (2008). I terroir di eccellenza qualitativa e internazionale del Sangiovese

in Toscana. Atti III Simp. Int. „Il Sangiovese‟ Firenze, Ed. ARSIA.

SCALABRELLI G., LAGOMARSINI G. (2008). Should we concern about the future of

disadvantaged viticulture areas with globalization?, Acta, 2nd Int. Cong. of Mountain And

Steep Slope Viticulture, CERVIM, Monforte de Lemos, Espagna, Feb. 13-15: 1-11.

211

SCALABRELLI G. (2009). Quale futuro per il vivaismo viticolo toscano?. I Georgofili,

Quaderni 2008 III Sezione Centro Ovest, Lo Sviluppo del vivaismo toscano, Pistoia

5/12/2008: 147-177, Felici editore.

SCALABRELLI G. (2009). Viticulture for the third millenium, safeguard and innovation.

Acta 16th

International GIESCO Symposium, July 12-15, 2009 University of California,

Davis: 63-70.

SCALABRELLI G. (2010). Non solo vino dal vigneto: agresto e altro. Vitenda N°16 (1):

136-137.

SCALABRELLI G. (2013). Wine and Food Excellence between Local identity and market

requirements. Wine and Food Tourism First European Conference. Ed. ETS: 87-120.

SCHREIER P., DRAWERT F., JUNKER A. (1976). Identification of volatile constituents

from grapes. J. Agric Food Chem. 24: 331-336.

SCIENZA A., BOGONI M., BRANCADORO L., VALENTI L., ROMANO F. (1990). La

conoscenza dei rapporti tra vitigno ed ambiente quale strumento programmatorio in

viticoltura: stima della vocazionalità viticola dell’Oltrepò Pavese. Vignevini, Suppl. al n.12,

4-62.

SCIENZA A. (1998). La zonazione viticola tra natura e cultura. In ‘Capire il territorio per

attuare una politica di qualità‟. Seminario Permanente Luigi Vigorelli, Bergamo: 31-38.

SCIENZA A. (2002). Viticoltura di precisione: programmare la qualità del vigneto.

Supplemento a „L‟informatore agrario‟.

SIMPSON R.F. (1978). 1,1,6-trimethyl-1,2-dihydronaphtalene: an important contributor to

the bottle aged bouquet of wine. Chem. Industr. 1:37.

SKOUROUMOUNISG. K., MASSY-WESTROPP R. A., SEFTON M. A., WILLIAMS P. J.

(1992). Tetrahed Lett. 33 (24) 3533.

STORCHI P., COSTANTINI., E.A.C., BUCELLI P. (2005). The Influence of Climate and

Soil on Viticultural and Enological Parameters of ‘Sangiovese’ Grapevines under Non-

irrigated Conditions. Proc. VIIth IS on Grapevine, Acta Hort. 689, ISHS: 333-340.

STORCHI P., ORLANDINI S., GALLIGANI U., PINZAUTI S., D‟ARCANGELO M.,

MANCINI M., DALLA MARTA A., MATTEOLI A. (2006). Valorizzazione del Sangiovese

attraverso lo studio dei rapporti tra ambiente e vitigno a scala aziendale: l’esempio della

fattoria ‘Poggio Casciano’. Atti Simp. Int. „Il Sangiovese identità e peculiarità di un vitigno

tipico e internazionale‟. Firenze, Ed. ARSIA: 387-394.

STRAUSS C. R., WILSON B., GOODLEY P. R., WILLIAMS P. J. (1986). Role of

monoterpenes in grape and wine PHENOL-FREE RIESLING GLYCOSIDES 247 flavor. In

212

Biogeneration of Aroma.(T. H. Parliament and R. Croteau, Eds.), pp. 222–242. Am. Chem.

Society, Washington.

STRAUSS C. R., WILSON B., WILLIAMS P. J. (1988) – J. Agric Food Chem. 36 569.

TAMAI G. (2009). Catalogo dei Cloni.Varietà di uva da vino. Edagricole, Calderini,

Bologna.

TOMASI D., CALO‟ A., PASCARELLA G., PITACCOA., BORSA D., GAIOTTI F. (2007).

Effetti dell’incremento termico sulla qualità dell’uva: il caso della Garganega e le sue forme

di allevamento. XXX Congresso mondiale della viticoltura e dell‟enologia Budapest 10-16

giugno.

TONINATO L., BERNAVA M., CRICCO J. BRANCADORO L. (2005). Caratterizzazione

dei terroirs di Vinci e Cerreto Guidi mediante le risposte del Sangiovese. L‟Informatore

Agrario, 2: 7-13.

TONINATO, L. BONCOMPAGNI, S. BRANCADORO, L. FAILLA, O. SCIENZA, A.

(2006). Zonazione e qualità dei vini nella provincia di Arezzo. L'Informatore Agrario (25):

54-57.

TURRI S. and INTRIERI C. (1987). Mappe isotermiche ed insediamenti viticoli in Emilia

Romagna. Vignevini, 10.

TURRI S. and INTRIERI C (1988). Fenologia e climatologia applicata alla Vitis vinifera:

appunti metodologici. ATTI, Accademia Italiana della Vite e del Vino, Siena, 40: 191-206.

UNIONE ITALIANA VINI. Laboratorio Chimico Sensoriale.

VAN LEEUWEN C., FRIANT P., CHONE‟ X.,TREGOAT O., KONDOURAS S.,

DUBORDIEU D. (2004). Influence of climate, soil, and Cultivar on terroir Am.

J.Enol.Vitic.55:3 207-217.

VAN LEEUWEN C., SEGUIN G. (2006). The concept of Terroir in Viticulture Journal of

Wine Research , Vol 17, No 1 1-10.

VAUDOUR E. (2002). The quality of Grapes and Wine in Relation to Geography: Notions of

Terroir at Various Scales. Journal of Wine Research Vol. 13 No 2 , 117-141.

VAUDOUR E. (2005). Grapevine responses to terroir: a gllobal approach J.Int.Sci.Vigna

Vin 39, n° 4, 149-162.

VAUDOUR E. (2005). I terroir. Definizioni, caratterizzazione e protezione, Edagricole,

Bologna.

VOIRIN S., BAUMES R., BITTEUR S., GUNATA Y., BAYANOVE C. (1990). Novel

monoterpene disaccharide glycosides of vitis vinifera grapes. J.Agric.Food Chem. 38:1373-

78.

213

VOUILLAMOZ F., IMAZIO S., STEFANINIM., SCIENZA A., GRANDO M.S. (2004).

Relazioni genetiche del Sangiovese. Atti II° Simp. Int. „Il Sangiovese identità e peculiarità di

un vitigno internazionale‟ (Abstract 22), Firenze 17-19 novembre.

WILLIAMS P., STRAUSS C., WILSON B., MASSY WESTROPP R. (1982). Novel

monoterpene disaccharide glycosides of vitis vinifera grapes and wines. Phytochem.,

21(8):2013-20.

WINTERHALTER P. (1993). In flavour precursors. Thermal and enzymatic conversions.

eds. R. Teranishi, G. Takeoka and M. Guntert. Acs series 490.

ACKNOWLEDGEMENTS TO:

Prof. Giancarlo Scalabrelli, Dr. Claudio D’Onofrio of Department of Agriculture, Food

and Environment, University of Pisa for viticultural advice and all the staff of „Laboratory of

Viticulture and Oenology‟ of San Piero a Grado.

Dott.ssa Angela Cuzzola of Department of Agriculture, Food and Environment, University

of Pisa for gas chromatography – mass spectrometry advice.

Dr. Giuseppe Ferroni of Department of Agriculture, Food and Environment, University of

Pisa for advice in the sensorial analysis of wines.

‘Beconcini’ estate, ‘Capannelle’ estate, ‘Casanova di Neri’ estate, ‘Castello di Albola’

estate, ‘ColleMassari’ estate, ‘Col d’Orcia’ estate, ‘Fattoria di Magliano’ estate, ‘La

Mannella’ estate and ‘Salustri’ estate for „Sangiovese‟ sampling.

Research carried out within the project „Qualitative characterization of grapevine Sangiovese

and wines in different area‟ supported by „Montecucco‟ and „Bertarelli‟ Foundations.

The qualitative characterization of Sangiovese grapevine ... · Ph.D thesis IN SCIENCE OF CROP...

Documents

Transcript of The qualitative characterization of Sangiovese grapevine ... · Ph.D thesis IN SCIENCE OF CROP...